JP2002354242A - Image processor, image reader, image forming device, and color copying machine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processor which prevents an image from looking unnatural in a switched boundary part due to switching of image processing. SOLUTION: The image processor which subjects inputted image data to prescribed processing and outputs the result is provided with a gray pixel detection part 323b and a gray pattern matching part 323h which detect muddle- density (gray) areas of the image represented by the picture data, a gray expansion part 323i which performs expansion processing of image data detected by the gray pattern matching part 323h, and a discrimination part 323m which detects the results of expansion processing in the gray expansion part 323i as non-character edge areas of the image, and areas expanded in the gray expansion part 323i are removed from the object of character discrimination. As a result, processing is switched by only middle-density areas, and areas having thick lines can be subjected to the same image processing independently of line width, and the image is prevented from looking unnatural in the switched boundary part due to switching of image processing.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for detecting characters in image data and performing appropriate image processing, an image reading apparatus having the image processing apparatus, and an image having the image processing apparatus. The present invention relates to a color copying apparatus including a forming apparatus and an image processing apparatus.

[0002]

2. Description of the Related Art Japanese Patent Laid-Open No.
The invention disclosed in 152,945 is known. According to the present invention, a white background is detected, and a character area having a small line width and a character area having a large line width are separately processed.

The invention disclosed in Japanese Patent Application Laid-Open No. 10-108012 is also known. According to the present invention, a white background is detected and a portion surrounded by the white background is processed as a picture region.

[0004]

However, when the image processing is switched based on the line width as in the former case, it seems unnatural because the processing is switched at the boundary of the line width to be switched. That is, for example, when a character having a low density of 0.7 to 1.0 is read and output, the character edge is emphasized, and the outline of the character is bordered, which may make the character hard to see.

An object of the present invention is to provide an image processing apparatus in which an image does not look unnatural at a boundary portion switched by switching image processing.

It is another object of the present invention to provide an image reading apparatus, an image forming apparatus, and a color copying apparatus using an image processing apparatus in which an image does not look unnatural at a boundary switched by switching image processing. It is in.

[0007]

To achieve the above object, a first means is an image processing apparatus which performs predetermined processing on input image data and outputs the processed image data. Medium density detection means for detecting a medium density area, expansion means for expanding the image data detected by the medium density detection means, and non-text for detecting the result of expansion processing by the expansion means as a non-character edge area of the image Edge determination means. The medium density area here corresponds to a gray area in the embodiment described later, and the medium density detection means corresponds to the gray pixel detection section 323b-1 and the gray pattern matching section 323h of the RGB white extraction section 323b. Note that the detection of the medium density region is described in detail in the section of <Gray Determination> described later.

A second means is the first means, wherein the expansion means performs an expansion processing after performing a contraction processing. This expansion processing is performed by the gray expansion section 323.
i.

A third means is the first means, wherein the medium density detection means detects the medium density area by pattern matching with a line pattern. This pattern matching is performed by the gray pattern matching unit 3
23h.

A fourth means is the first means, further comprising an edge detecting means for detecting an edge. In addition, the edge detecting means is the edge extracting unit 3
22.

The fifth means includes an image processing apparatus according to the first to fourth means and an image reading means for inputting image data generated by separating and reading an original image to the image processing apparatus. A reading device is configured. Note that the image reading unit corresponds to the scanner 200 in an embodiment described later.

The sixth means is an image processing apparatus according to the first to fourth means, and forms an image based on image data output from the image processing apparatus, and forms the formed image on paper. The image forming apparatus comprises an image output means for outputting an image. In an embodiment described later, the image output unit corresponds to the printer 400.

The seventh means is an image processing apparatus according to the first to fourth means, an image reading means for inputting image data generated by reading an original image by color separation to the image processing apparatus, An image is formed based on the image data output from the image processing apparatus, the formed image is formed on a sheet, and an image output means for outputting the image forms a color copying apparatus.

An eighth means is the seventh means, further comprising control means for analyzing an external print instruction command and causing the image output means to print out external image information. I do. In the embodiment described later, the control means corresponds to the printer controller 16.

[0015]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an image processing apparatus according to the present invention will be described in detail with reference to the drawings.

<Digital Full-Color Copying Machine> FIG. 1 shows an outline of a mechanism according to an embodiment of the present invention. FIG. 1 is a schematic configuration diagram of a digital full-color copying machine including an image processing apparatus according to the present embodiment.

The digital full-color copying machine according to this embodiment includes a color image reading device (hereinafter, referred to as a scanner) 200 and a color image recording device (hereinafter, referred to as a color printer) 400.

The scanner 200 includes a contact glass 20
2 is formed on the color sensor 207 via the illumination lamp 205, the mirror groups 204A, 204B, 204C, and the lens 206, and color image information of the document is, for example, blue (hereinafter, referred to as B). ), Green (hereinafter, referred to as G) and red (hereinafter, referred to as R)
Is read for each color separation light and is converted into an electric image signal. The color sensor 207 has three lines CC in this example.
It is composed of a D sensor and reads B, G, and R images for each color. Based on the B, G, and R color separation image signal intensity levels obtained by the scanner 200, color conversion processing is performed by an image processing unit (not shown), and black (hereinafter, referred to as B)
k), cyan (hereinafter, referred to as C), magenta (hereinafter, referred to as M) and yellow (hereinafter, referred to as Y) color image data including recording color information.

Using the color image data, Bk, C, M, and Y images are formed on an intermediate transfer belt by a color printer 400, and then transferred to transfer paper. The scanner 200 receives a scanner start signal synchronized with the operation of the color printer 400, and scans a document in the left arrow direction by an illumination / mirror optical system including an illumination lamp 205 and mirror groups 204A, 204B, and 204C. Image data of one color is obtained for each scan. Each time, while sequentially visualizing the color printer 400,
These are superposed on the intermediate transfer belt to form four full-color images.

A writing optical unit 401 as an exposure unit of the color printer 400 converts color image data from the scanner 200 into an optical signal, performs optical writing corresponding to a document image, and forms an electrostatic image on the photosensitive drum 414. Form a latent image. The optical writing optical unit 401 includes a laser light emitter 441, a light emission drive control unit (not shown) for driving the light emission, a polygon mirror 443, a rotation motor 444 for rotating the same, a fθ lens 442, a reflection mirror 446, and the like. Have been. Photoconductor drum 414
Rotates counterclockwise in the figure as indicated by the arrow, but around it, a photoconductor cleaning unit 421, a neutralization lamp 414M, a charger 419, a potential sensor 414D for detecting a latent image potential on the photoconductor drum, A developing unit selected from the revolver developing device 420, a developing density pattern detector 414P, an intermediate transfer belt 415, and the like are arranged.

The revolver developing device 420 includes a BK developing device 4
20K, C developing unit 420C, M developing unit 420M, Y developing unit 420Y, and a revolver rotation drive unit (not shown) that rotates each developing unit in the counterclockwise direction as shown by the arrow. In order to visualize the electrostatic latent image, each of these developing units makes the ears of the developer contact the surface of the photosensitive drum 414, and rotates the developing sleeves 420KS, 420CS, and 4CS.
20 MS, 420 YS, and a developing paddle that rotates to assemble and stir the developer. In the standby state, the revolver developing device 420 is the BK developing device 420
When the copying operation is started, the scanner 200 starts reading BK image data at a predetermined timing, and based on the image data, writes the BK image data using a laser beam and performs latent image writing. Formation begins. Hereinafter, the electrostatic latent image based on the Bk image data is referred to as a Bk latent image. The same applies to each of the C, M, and Y image data. In order to enable development from the tip of this Bk latent image, Bk
Before the leading end of the latent image reaches the developing position of the developing device 420K, the rotation of the developing sleeve 420KS is started to develop the Bk latent image with Bk toner. Thereafter, the developing operation of the Bk latent image area is continued. When the rear end of the latent image passes the Bk latent image position, the developing operation of the next color developing unit is immediately performed from the developing position of the Bk developing unit 420K. The revolver developing device 420 is driven and rotated to the position. This rotation operation is completed at least before the leading end of the latent image based on the next image data arrives.

When the image forming cycle is started, the photosensitive drum 414 rotates counterclockwise as indicated by an arrow,
The intermediate transfer belt 415 is driven by a drive motor (not shown).
Rotate clockwise. With the rotation of the intermediate transfer belt 415, BK toner image formation, C toner image formation, M toner image formation, and Y toner image formation are sequentially performed, and finally, the intermediate transfer is performed in the order of BK, C, M, and Y A toner image is formed over the belt 415. The formation of the BK image is performed as follows. In other words, the charger 419 causes the photoconductor drum 414 to have a negative charge of about -70 by corona discharge.
It is uniformly charged to 0V. Subsequently, the laser diode 44
1 performs raster exposure based on the Bk signal. When the raster image is exposed in this manner, in the initially exposed portion of the photosensitive drum 414 that is uniformly charged, the charge proportional to the amount of exposure light disappears, and an electrostatic latent image is formed.
The toner in the revolver developing device 420 is negatively charged by stirring with the ferrite carrier, and the BK developing sleeve 420KS of the present developing device is
A power supply circuit (not shown) biases the 14 metal base layers to a potential at which a negative DC potential and an AC are superimposed. As a result, the toner does not adhere to the portion of the photosensitive drum 414 where the charge remains, and the Bk toner is adsorbed to the portion having no charge, that is, the exposed portion, and the Bk visible light similar to the latent image is obtained. An image is formed. The intermediate transfer belt 415 includes a driving roller 415D, a transfer opposing roller 4
15T, the cleaning opposing roller 415C, and the driven roller 415F are stretched around and driven to rotate by a drive motor (not shown). The Bk toner image formed on the photoconductor drum 414 is applied to the surface of the intermediate transfer belt 415 that is driven at a constant speed in a state of contact with the photoconductor, by a belt transfer corona discharger (hereinafter, referred to as a belt transfer unit) 416. Is transcribed. Hereinafter, the transfer of the toner image from the photosensitive drum 414 to the intermediate transfer belt 415 is referred to as belt transfer. Some untransferred residual toner on the photoconductor drum 414 is cleaned by the photoconductor cleaning unit 421 in preparation for reuse of the photoconductor drum 414. The collected toner is stored in a waste toner tank (not shown) via a collection pipe.

On the intermediate transfer belt 415, Bk, C, M, and Y toner images sequentially formed on the photosensitive drum 414 are sequentially aligned on the same surface, and a four-color superimposed belt transfer image is formed. After that, collective transfer is performed on transfer paper by a corona discharge transfer device. On the photosensitive drum 414 side, the process proceeds to the C image forming process after the BK image forming process.
The reading of the C image data by 0 starts, and the formation of the C latent image is performed by writing the laser light with the image data.
The C developing device 420C moves the Bk
After the rear end of the latent image has passed and before the front end of the C latent image has arrived, the rotating operation of the revolver developing device is performed to develop the C latent image with the C toner. Thereafter, development of the C latent image area is continued, but when the rear end of the latent image has passed, the revolver developing device 420 is driven in the same manner as in the case of the Bk developing device, and the C developing device 420C is sent out. Of the M developing device 420M is positioned at the developing position. This operation is also performed before the leading end of the next M latent image reaches the developing unit. In the process of forming the M and Y images, the operations of reading the image data, forming the latent image, and developing are the same as those of the Bk image and the C image described above, and a description thereof will be omitted.

The belt cleaning device 415U includes an inlet seal, a rubber blade, a discharge coil, and a mechanism for contacting and separating the inlet seal and the rubber blade. 1
After the Bk image of the color is transferred to the belt, while the images of the second, third, and fourth colors are transferred to the belt, the entrance seal, the rubber blade, and the like are separated from the intermediate transfer belt surface by the blade contact / separation mechanism. .

A paper transfer corona discharger (hereinafter, referred to as a paper transfer device) 417 uses an AC corona discharge method to transfer the superimposed toner image on the intermediate transfer belt 415 to transfer paper.
+ DC or a DC component is applied to the transfer paper and the intermediate transfer belt.

Transfer paper of various sizes is stored in a transfer paper cassette 482 in the paper feed bank.
The sheet 83 is fed and conveyed in the direction of the registration roller pair 418R. Reference numeral 412B2 indicates a paper feed tray for manually feeding OHP paper, thick paper, and the like. At the timing when the image formation is started, the transfer paper is fed from any one of the paper feed trays, and is waiting at the nip portion of the registration roller pair 418R. Then, when the leading end of the toner image on the intermediate transfer belt 415 approaches the paper transfer unit 417, the registration roller pair 418R is driven so that the leading end of the transfer paper coincides with the leading end of this image, and the transfer of the paper and the image is performed. Matching is performed. In this way, the transfer paper is superimposed on the color superimposed image on the intermediate transfer belt and passes over the paper transfer unit 417 connected to the positive potential. At this time, the transfer paper is charged with a positive charge by the corona discharge current, and most of the toner image is transferred onto the transfer paper. Subsequently, the transfer paper is discharged when passing through a separation static eliminator by a static elimination brush (not shown) disposed on the left side of the paper transfer device 417, and the intermediate transfer belt 4
15 and is transferred to the paper transport belt 422. The transfer paper on which the four-color superimposed toner image is collectively transferred from the intermediate transfer belt surface is transported to the fixing device 423 by the paper transport belt 422, and
The toner image is melted and fixed at the nip between the fixing roller 423A and the pressure roller 423B controlled to a predetermined temperature, sent out of the main body by a discharge roll pair 424, and stacked face up on a copy tray (not shown).

The photosensitive drum 414 after belt transfer
The surface is cleaned by a photoconductor cleaning unit 421 including a brush roller, a rubber blade, and the like.
Further, the charge is uniformly removed by the charge removing lamp 414M. Further, the intermediate transfer belt 415 after transferring the toner image to the transfer paper
Again, the blade is pressed by the blade contact / separation mechanism of the cleaning unit 415U to clean the surface.
In the case of the repeat copy, the operation of the scanner and the image formation on the photoconductor are continued from the first-color image process on the first sheet, and then proceed to the first-color image process on the second sheet at a predetermined timing. In the case of the intermediate transfer belt 415, the second Bk toner image is belt-transferred to an area whose surface has been cleaned by a belt cleaning device, following a batch transfer process of the first four-color superimposed image onto transfer paper. So that Thereafter, the operation is the same as that of the first sheet.

The digital full-color copying machine shown in FIG.
When print data is provided from a host such as a personal computer through a LAN or a parallel I / F, the print data can be printed out (image output) by the color printer 400, and the image data read by the scanner 200 can be transmitted to a remote facsimile. This is a color copier with a composite function that can print out image data to be transmitted to and received from the printer. This copier is connected to a public telephone network via a private branch exchange PBX, and can communicate with a facsimile communication and a management server of a service center via the public telephone network.

<Electrical System> {Outline of System} FIG. 2 shows an outline of an electric system of the digital full-color copying machine shown in FIG. FIG. 2 illustrates a control device of a digital full-color copying machine with a focus on the main controller 10. The main controller 10 controls the entire copying machine. Main controller 1
0, control of the operation / display board OPB for performing display to the operator and input of function setting from the operator, control of the editor 15, the scanner 200 and the optional ADF, control of writing a document image in the image memory, and
A scanner controller 12, a printer controller 16, an image processing unit (IPU) 300, and a color printer 40 for performing control and the like for forming an image from an image memory
A distributed control device such as an engine controller 13 which controls an image forming engine which is in the area 0 and performs charge, exposure, development, paper feed, transfer, fixing, and transfer of a transfer sheet is connected. Each decentralized control device and the main controller 10 exchange machine status and operation commands as needed. Further, a main motor and various clutches necessary for paper conveyance and the like are also connected to a driver (not shown) in the main controller 10. Reference numeral 11 denotes an IC card, and reference numeral 14 denotes a sorter controller. The IC card 11 is used, for example, for managing the number of copies for each department.

The color printer 400 includes an electric circuit that drives a mechanism element that performs charging of the photosensitive member 414, image exposure by a laser writing unit, development, transfer, fixing, and paper discharge, starting with paper supply from a paper supply tray. And a control circuit, various sensors, and the like.

The printer controller 16 analyzes an external image such as a personal computer or the like and a command for instructing printing, develops a bitmap into a printable state as image data, and drives the printer 400 via the main controller 10 to generate an image. Print out the data. A LAN control 19 and a parallel I / F 18 are provided to receive and operate images and commands via the LAN and the parallel I / F.

When a facsimile transmission instruction is given, the facsimile controller 17 drives the scanner 200 and the IPU 300 via the main controller 10 to read the image of the document and to transmit the image data to the facsimile communication line via the communication control 20 and the PBX. Send out. Upon receiving a facsimile call from the communication line and receiving image data, the printer 4
00 to print out the image data.

FIG. 3 shows an image processing unit (IPU) 30.
0 is shown. In the figure, R, G, and B image data output from the scanner 200 is provided to the IPU 300 via the interface (1) 351. When the BR unit 355 instructs recording of B or R single color, selection and aggregation of R, G, and B image data are performed, but description of image recording processing in this mode is omitted.
The R, G, B image data given to the IPU 300 is R
The GBγ correction unit 310 converts the reflectance data (R, G, B data) into density data (R, G, B data).

The document recognizing section 320 detects the density R, G, B
Based on the data, it is determined whether the image area to which the data is addressed is a character area (character or line drawing area) or a picture area (a picture or picture area and not a character area), and the C / P signal and the B / The C signal is provided to the main controller 10 via the RGB filter unit 330 and the interface (3) 353. Note that the C / P signal and B /
The C signal is a C / P signal: a 1-bit signal, where 1 indicates a character edge area and 0 indicates a picture area.

B / C signal: 1-bit signal, H
("1") indicates an achromatic region, and L ("0") indicates a chromatic region.

{Document Recognition Unit 320 (FIG. 4)} FIG. 4 shows the functions of the document recognition unit 320 in block divisions. The document recognizing unit 320 performs character edge detection, picture detection, and chromatic / achromatic detection to generate a C / P signal representing a character edge area or a picture area and a B / C signal representing a chromatic area / achromatic area. appear.

The document recognizing section 320 is roughly divided into a filter section 321, an edge extracting section 322, and a white area extracting section 32.
3, a halftone dot extracting section 324, a color determining section 325 and a comprehensive determining section 326. Here, a case where the reading density of the scanner 200 is about 600 dpi will be described as an example.

{Filter Unit 321} Filter Unit 321
Corrects the G image data generated by the scanner 200 mainly for extracting the edges of the character. Here, the scanner 2
Since the data read at 00 may be blurred due to the performance of the lens or the like, an edge emphasis filter is applied. However, here, it is necessary to simply emphasize the edge on the document and not to emphasize the line pattern for gradation expression widely used in copying machines. If the line pattern is emphasized, a picture (a gradation expression area by the line pattern) is extracted as an edge, and there is a possibility that the character line may be erroneously determined as a character edge. is there. Also, as shown in FIG. 8, a 600 dpi line pattern A
And 400 dpi line pattern B have different repetition periods, so it is difficult to avoid emphasis with the same filter coefficient. Therefore, by detecting the period of the image pattern,
Switch filter coefficients. In FIG. 8, the sum of the width of one white block and the width of one black block in contact with it in the main scanning direction x is the line pitch (constant width: a predetermined number of pixels), that is, the line period, and the low density halftone In the case of, the white block width increases and the black block width decreases. As the density becomes higher, the white block width becomes narrower and the black block width becomes wider.

In this embodiment, the pixel matrix of the filter unit 321 is defined as 7 pixels in the main scanning direction x 5 pixels in the sub-scanning direction y (mechanical original scanning direction of the scanner 200) in FIG. As shown in the block of the filter unit 321, each weighting coefficient a1 to a7, b1 is assigned to each pixel.
There are two sets of coefficient groups (coefficient matrices) A and B addressed to .about.b7, c1 to c7, d1 to d7, and e1 to e7.
The next coefficient group A is a filter processing coefficient that suppresses the emphasis of the 600 dpi line pattern A in FIG. 8 and also emphasizes the edges of characters.
This is a filter processing coefficient that suppresses the emphasis of the 400 dpi line pattern B and emphasizes the edges of characters.

Coefficient group A 0-10-20-10-10-10-20-10-10-1-20-1-10-10-10-20-10-10-1 0-20-10.

Coefficient group B-100-200--1-100-200--1-100-120-110--100--20--100 0-200-1.

The horizontal direction is a line in the main scanning direction x, and the vertical direction is a line in the sub-scanning direction y. Coefficient groups A and B
Of the first row in the group is the filter 32 in FIG.
Coefficients a1 to a1 in the first row of the coefficient matrix of one block
a7, and “2” at the center of the third row of the coefficient groups A and B.
“0” is the coefficient of the pixel at the center of the third row c1 to c7 of the coefficient matrix of the block of the filter unit 321, that is, the coefficient c4 of the pixel of interest. The sum (product sum value) of the products (total 7 × 5 = 35) obtained by multiplying each coefficient of the coefficient matrix by the value represented by the image data of the pixel addressed thereto is calculated as the target pixel (c4
(Pixels to which is assigned) is supplied to the edge extraction unit 322 and the white region extraction unit 323 as image data values processed by the filter unit 321. Here, the target pixel is a pixel to be currently processed, and is sequentially updated to a pixel whose position is different in the x direction and the y direction.

The coefficient group A has a capacity of 600 dp shown in FIG.
A negative coefficient (a coefficient having a small value) is distributed at the line pitch of the line pattern A of i, and 0 (a coefficient having a somewhat large value) is distributed between them. (Extremely large coefficient). Accordingly, when the image data (pixel of interest) is the black / white edge of the area of the line pattern A, the weighted average value (product sum value) derived therefrom is a character edge that is not the line pattern A. It is much lower than at some point.

The coefficient group B is 400 dp shown in FIG.
A negative coefficient (a coefficient having a small value) is distributed at the line pitch of the line pattern B of i, and 0 (a coefficient having a relatively large value) is distributed between them. Is assigned 20 (an extremely large coefficient). Thus, when the image data (pixel of interest) is a black / white edge of the area of the line pattern B, the weighted average value (product sum value) derived therefrom is a character edge that is not the line pattern B. It is much lower than at some point.

In the filter section 321, when one of the following conditions 1 and 2 is satisfied, that is, 400 in FIG.
When there is a high possibility that the line pattern B is the dpi line pattern B, the filter processing is performed by the coefficient group B. When not, the filter processing is performed by the coefficient group A. (Condition to see if (white section in FIG. 8)) (D [3] [1] <D [3] [2]) & (D [3] [7] <D [3] [6]) & (ABS (D [3] [2] -D [3] [4])> ABS
(D [3] [4] -D [3] [1])) & (ABS (D [3] [6] -D [3] [4])> ABS
(D [3] [4] -D [3] [7]))-Condition 2- [Condition to see if the 400 dpi line pattern B is dark (black section in FIG. 8)] (D [3] ] [1]> D [3] [2]) & (D [3] [7]> D [3] [6]) & (ABS (D [3] [2] -D [3] [4] )> AB
S (D [3] [4] -D [3] [1])) & (ABS (D [3] [6] -D [3] [4])> ABS
(D [3] [4] -D [3] [7]) Here, D [i] [j] is a pixel at a position of x = i, y = j on a pixel matrix of x, y distribution. , For example, D [3] [1] is the coefficient a of the coefficient matrix shown in the block of the filter unit 321 in FIG.
3 is a value represented by the image data of the addressed pixel.
"&" Means "AND" and "ABS"
Absolute value operator. The target pixel is D [4] [3]
It is.

When the above condition 1 or 2 is satisfied, the pixel of interest at that time is 4 pixels at the time of reading at 600 dpi shown in FIG.
Assuming that the image belongs to the area of the line pattern B of 00 dpi, the filter processing of the character edge emphasis is performed using the coefficient group B. If neither of the conditions 1 and 2 is satisfied, 600 dpi at the time of reading 600 dpi shown in FIG.
The character edge enhancement filter processing is performed using the coefficient group A for preventing the line pattern A from being emphasized. That is, the image cycle (pitch) is detected so that the image pattern of the specific cycle is not emphasized. It is possible to emphasize the edge of the character without emphasizing the line pattern.
Note that FIG. 4 shows a mode in which the G image data is referred to for the edge processing. It can be applied to signals that express dark and light.

<Edge Extraction> {Edge Extraction Unit 322} The character area has many high-level density pixels and low-level density pixels (hereinafter, referred to as black pixels and white pixels). Black pixels and white pixels are continuous. The edge extraction unit 322 detects a character edge based on the continuity of each of such black pixels and white pixels.

{Ternary conversion section 322a}
In step 2a, the filter unit 321 performs G edge filtering by using two types of thresholds TH1 and TH2.
The image data (input data of the edge extraction unit 322) is ternarized. For example, when the image data represents 256 gradations (0 = white) from 0 to 255, the thresholds TH1 and TH2 are set to, for example, TH1 = 20 and TH2 = 80. 3
If the input data <TH1, the value conversion unit 322a indicates that the pixel addressed to the data is a white pixel.
The input data is converted into ternary data. If TH1 ≦ input data <TH2, the input data is converted into ternary data representing a halftone pixel. If TH2 ≦ input data, ternary data representing a black pixel. Convert the input data to

{Continuous black pixel detecting section 322b, continuous white pixel detecting section 322c} The continuous black pixel detecting section 322b and the continuous white pixel detecting section 322c determine the position where black pixels continue and the white pixel Successive locations are detected by pattern matching. In this embodiment, in this embodiment, the patterns BPa to BPd and WPa to BPd of the 3 × 3 pixel matrix shown in FIG.
WPd is used. In the pattern shown in FIG. 9, a black circle indicates the above-described black pixel, a white circle indicates the above-described white pixel, and blank pixels without any circle indicate black pixels, halftone pixels, and white pixels. It does not matter which one of them. The pixel at the center of the 3 × 3 pixel matrix is the pixel of interest.

The black pixel continuation detecting section 322b determines that the distribution of the contents of the ternary data is the black pixel distribution pattern BP shown in FIG.
When matching is performed with any of a to BPd, the pixel of interest at that time is set as a “black continuous pixel” and data representing the pixel is given to the pixel of interest. Similarly, the white pixel continuation detection unit 322c
When matching with any of the white pixel distribution patterns WPa to WPd shown in FIG. 9, the target pixel at that time is set as a “white continuous pixel”, and data representing the pixel is given to the target pixel.

{Neighboring Pixel Detection Unit 322d} The next neighboring pixel detection unit 322d determines the detection results of the black pixel continuation detection unit 322b and the white pixel continuation detection unit 322c in the vicinity of the pixel of interest in the neighborhood pixel detection unit 322d. By examining whether there is a black continuous pixel or a white continuous pixel, it is determined whether the target pixel is in an edge area or a non-edge area. More specifically, in the present embodiment, when one or more black continuous pixels and one or more white continuous pixels are present in a 5 × 5 pixel matrix block, the block is defined as an edge region. If not, the block is determined to be a non-edge area.

{Isolated point removing unit 322e} Further, since the character edge exists continuously, the isolated point removing unit 322e corrects the isolated edge to a non-edge area. Then, an edge signal of “1” (edge area) is output to a pixel determined to be an edge area, and an edge signal of “0” (non-edge area) is output to a pixel determined to be a non-edge area.
Output a signal.

<White Area Extraction> {White Area Extraction Unit 323} White Area Extraction Unit 32 in FIG.
Reference numeral 3 denotes a binarization unit 323a, an RGB white extraction unit 323b, a white determination unit 323c, a white pattern matching unit 323d, a white pattern correction unit 323j, a white expansion unit 323h, a white contraction unit 323l, and a white correction unit as shown in FIG. Section 323g, gray pattern matching section 323h, gray expansion section 323
i and a determination unit 323m. It should be noted that the white area extraction unit 323 in FIG. 4 is obtained by replacing the part M in FIG.

{Binarization section 323a} Binarization section 323a
Converts the edge-enhanced output of the image density data (G image data) of the filter unit 321 into a binary value with a threshold thwsb, and outputs the white data of the white data referred to by the white pattern matching unit 323d (Step SS7 in FIG. 5 representing the processing). A binarized white determination signal for generation is generated. In this embodiment, the edge enhancement output is 256 gradations from 0 to 255, 0 is white without density, an example of the threshold thwsb is 50, and the value of the edge enhancement output is thwsb = 50.
If it is smaller, the binarization unit 323a determines "binary white" and generates a binary white determination signal "1". When the value of the edge emphasis output is greater than thwsb = 50, a binary white determination signal “0” is generated.

{RGB white extraction unit 323b} The RGB white extraction unit 323b performs 1) RGB white background detection 2) color background detection 3) gray pixel detection to determine whether the image data is a white area or a gray area (medium density area). Determine whether or not.

1) RGB white background detection In the RGB white background detection, a white background separation operation is activated by detecting a white background region from R, G, B image data. That is, the processing of white background separation is started. Specifically, as shown in the pattern WBP of FIG.
If all of the R, G, and B image data of the pixel matrix is smaller than the threshold thwss, the target pixel (the center pixel of the 3 × 3 pixel matrix) is determined to be a white region, and the process of the white pattern matching unit 323d (FIG. Active (“1”) in step S3 of FIG.
To This is to detect whether there is a white pixel area having a certain extent. In this embodiment, each of the R, G, and B image data is also 256 from 0 to 255 in this embodiment.
Is a gradation, 0 is a base level without density, and a threshold value t
hwss <thwsb, and an example of thwss is 4
0, and all of the R, G, B image data is thws
If s is smaller than 40, it is determined as "white background" and a white background determination signal "1" is generated. When any of the R, G, and B image data has a threshold value of thwss = 40 or more, a white background determination signal “0” is generated.

2) Color ground detection Color ground is detected so that a light color is not determined as a white background. A. Here, first, the sign of each pixel of the 5 × 5 pixel matrix centering on the target pixel is represented by the pattern MPp in FIG.
, The central pixel c3 (M
If the RGB difference (the difference between the maximum value and the minimum value of the R, G, and B image data addressed to one pixel) of the Ca-MCd X mark pixel is larger than the threshold thc, the color pixel determination signal a is set to “1” (color). "0" (black and white pixel) when the value is equal to or less than the threshold thc.
And B. A peripheral pixel group on one side of the target pixel 画素 (MCa to MCa in FIG. 11)
If the R, G, and B image data of any of the pixels (in MCd) are all equal to or smaller than the threshold thwc, the one-side white determination signal b is set to “1” (white pixel). (Non-white pixel). The threshold value thwc is, for example, 2
0. C. Peripheral pixel group □ on the other side of the target pixel (MCa in FIG. 11)
If the R, G, and B image data of any of the pixels (from among .about.MCd) are all equal to or smaller than the threshold value thwc, the other-side white determination signal c is set to “1” (white pixel). (Non-white pixel). D. In any of the patterns MCa to MCd in FIG. 11, a AND (exclusive NOR of b and c) =
When “1” is established, that is, when a = “1” (the target pixel is a color pixel) and b and c match (white pixels on both sides of the target pixel or non-white pixels on both sides), the target The color ground determination signal d addressed to the pixel is set to “1” (color ground). The color ground determination signal d is referred to by the white pattern matching unit 323d (Step S6 in FIG. 10 representing the processing).

The above-described pattern matching A to D is performed so that when the surroundings of a black character are slightly colored due to an RGB reading position shift, the color is not picked up as a color. At the colored position around the black character, (exclusive NOR of b and c) or ゛ “0” (one of both sides of the target pixel is a white pixel and the other is a non-white pixel). In this case, the color ground determination signal d = It becomes "0" (non-colored ground). In addition, when the pixel of interest is a color pixel whose periphery is surrounded by a white background, the color ground determination signal d = “1” (color background), and even when a line is complicated, a light color pixel is detected as a color background. be able to. That is, where a line is complicated, a white place is not originally read completely white, but the processing A. RG
If the B difference is small, it is not determined that the pixel is a color pixel.
hwc is set more strictly than the white background where the density is to be checked (for example, thwss = 40, thwsb = 50, but thwc = 20). Can be accurately detected as a color background.

It should be noted that there may be a case where valley white pixels are detected during the color detection. In the valley white pixel detection, a valley white pixel in a small white area that cannot be detected by the RGB white background detection is indicated by G
Detection is performed based on the 5 × 5 pixel matrix distributions RDPa and RDPb of the image data. Specifically, based on the 5 × 5 pixel matrix distribution RDPa, miny = min (G [1] [2], G [1] [3],
G [1] [4], G [5] [2], G [5] [3], G
[5] [4]) is calculated. That is, in the 5 × 5 pixel matrix distribution RDPa shown in FIG.
iny is extracted. Then, maxy = max (G [3] [2], G [3] [3],
G [3] [4]) is calculated. That is, in the 5 × 5 pixel matrix distribution RDPa shown in FIG.
Extract axy. Next, mint = min (G [2] [1], G [3] [1],
G [4] [1], G [2] [5], G [3] [5], G
[4] [5]) is calculated. That is, the lowest density mint in the group of pixels with black circles is extracted from another 5 × 5 pixel matrix distribution RDPb shown in FIG. Then, maxt = max (G [2] [3], G [3] [3],
G [4] [3]) is calculated. That is, in the 5 × 5 pixel matrix distribution RDPb shown in FIG.
axt is removed. Here, min () is a function for detecting the minimum value. max () is a function for detecting the maximum value. Next, OUT = ((miny-maxy)> 0) #
((Mint-maxt)> 0) is calculated. That is, (miny-maxy) and (min
Of the t-maxt), the larger positive value is the valley detection value OUT. If the value of OUT is equal to or greater than a certain threshold, the target pixel (the center pixel of RDPa or RDPb) is regarded as a valley white pixel. To detect. As described above, the valley state of the image is detected, and the 1) RGB white background detection compensates for a portion that is difficult to detect.

{White determination section 323c} Here, the state variables MS and SS [I] used for white determination are updated. The contents are shown in the flowchart of FIG. Here, the state variable M
S is addressed to the pixel of the processing target line (line of interest),
The state variable SS [I] is addressed to the pixel one line before (the processed line) of the line to be processed, and is 4-bit white background information indicating the degree of white on the white background. Is generated by
The highest value represented by the state variables MS and SS [I] is 15
, Which means the whitest degree, and the lowest value is 0. That is, the state variables MS and SS [I] are:
This is data indicating the degree of white, and the larger the value, the stronger white. At the start of the copying operation, the state variable M
S and SS [I] are both initialized to zero.

In the process shown in FIG. 5, first, the state variable one line before the target pixel to be processed, that is, the white background information S
S [I] is compared with the state variable of the pixel one pixel before (the preceding pixel: processed pixel) on the same line as the target pixel, that is, the white background information MS (step S1), and the white background information SS [ I] is larger than the temporary white background information MS of the target pixel (step S2). Otherwise, the state variable MS of the preceding pixel is set to the temporary white background information M of the target pixel.
S. This means that information closer to white among the white background information of the peripheral pixels is selected.

After the copying operation is started, 1) RGB
When a white area, that is, a white background is detected by the white background detection [1)
White background determination signal output from RGB white background detection = “1”],
The white background information SS [I] of the pixel one line before the target pixel is set to 1
5 (steps S3 and S4), and the white background information MS of the target pixel is also set to 15 (step S5). Then, the white background information MS of the pixel of interest is written at the main scanning position (F) of the pixel of interest in the line memory for the current line (line of interest) of the line memory LMP shown in FIG. SS [I] is the line memory LMP shown in FIG.
Is written to the main scanning position (F) of the pixel of interest in the line memory for one line before (steps S3, S4, S5). Next, the white background information SS [I] addressed to the pixel of the previous line is propagated to the pixel of the previous line as follows (steps S14 to S14).
17). [I] means the main scanning position of the pixel of interest, and [I-1] means the position of the pixel one pixel before (the pixel immediately before the pixel of interest) in the main scanning direction x.

When SS [I-1] <SS [I] -1, SS [I-1] = SS [I] -1 is set in the line memory (steps S14, S1).
5). That is, in the line one line before the pixel of interest, the white background at the position (F) of the pixel of interest is located ahead of the white background information SS [I-1] one pixel before (E) the pixel (F) in the main scanning direction. The value “SS” obtained by subtracting 1 from the information SS [I]
If [I] -1 ”is larger (white is stronger), the white background information SS [I-1] addressed to the pixel (E) one pixel before the position (F) of the pixel of interest on the previous line. Is updated to a value obtained by lowering the white intensity by 1 from the white background information SS [I] at the position (F) of the target pixel.

Next, when SS [I-2] <SS [I] -2, SS [I-2] = SS [I] -2 is set in the line memory (steps S16 and S17-).
Next, when SS [I-3] <SS [I] -3, SS [I-3] = SS [I] -3 is set in the line memory (steps S16 and 17-).
14, 15).

Similarly, when SS [I-15] <SS [I] -15, SS [I-15] = SS [I] -15 is finally set in the line memory (step S16, 17-
14, 15). The lower limit value MIN of the value of the white background information SS [I] is 0, and when it is less than 0, it is kept at 0. This is the same in step S13 described later.

By the processing in these steps 14 to 17, the white background information SS one line before and before the main scanning position of the target pixel is replaced with the white background information MS of the target pixel, and then the positional deviation of one pixel in the main scanning direction x. , And the white background information of the pixel of interest propagates at the aforementioned reduction rate in the main scanning direction x one line before and behind the main scanning (white propagation processing). However, this is a case where the value of the white background information one line before is smaller. For example, the pixel before one line is the one. If the white background (white area) is detected by the RGB white background detection, the white background information is 15 which is the highest value, and no rewriting is performed.

When the target pixel is updated and becomes a non-white background [1) White background determination signal output from RGB white background detection = “0”], the process proceeds from step S3 to step S6 and subsequent steps, where the target pixel is Not the ground [the above [2] color ground detection signal d = “1”] which is the output of the color ground detection (it is a non-color ground), and binarized white [the output 2 of the binarization unit 323a.
When the value-determined white determination signal is “1”] and the state variable of the pixel of interest temporarily determined in Steps 1 and 2, that is, the white background information MS is equal to or greater than the threshold thw1 (for example, 13), the white background addressed to the pixel of interest is The information MS is incremented by 1 (steps S6 to S10). That is, the white value is updated to a value that is strong by one. The maximum value max of the white background information MS is set to 15, and when it exceeds 15, it is kept at 15 (step S
9, 10). The steps S5 and S14 to S17 are also executed when the vehicle has proceeded along this route. That is, white propagation processing is performed.

Although the target pixel has a non-colored background and a binary white, the white background information MS is smaller than thw1 (for example, 7).
When w2 (for example, 1) or more and a valley white pixel,
The state variable MS is kept as it is (step S
8, 11, 12). The steps S5 and S14 to S17 are also executed when the vehicle has proceeded along this route. That is, white propagation processing is performed.

If none of the above conditions is met, that is, if the target pixel is a color background or non-binarized white, the white background information MS of the target pixel is decremented by one (step S13). That is, the information is updated to white background information in which the degree of white is weak by one. The minimum value MIN of the white background information MS is 0, and when it is less than 0, it is kept at 0. The steps S5 and S14 to S17 are also executed when the vehicle has proceeded along this route. That is, white propagation processing is performed.

By the generation of the white background information MS, the white information can be transmitted to the peripheral pixels via the state variable (white background information) MS on the line memory LMP. The generation of the white background information MS is based on the color data (R, R,
Generation of white background information MS corresponding to the colors of the system of steps S3-4-5-14 to 17 in FIG. 5 based on the RGB white background determination signal indicating that the white background is present when all of the G and B image data are smaller than the threshold thwss = 40. And an edge enhancement output of the density data (G image data) (the filter unit 321).
Is smaller than the threshold value thwsb = 50, the steps S7 to S7 in FIG.
13-5-14 to 17 white background information MS corresponding to density
Including the generation of

The white determination unit 323c first generates a white background determination signal "1" in the 1) RGB white background detection in the RGB white extraction unit 323b until a white region is detected, that is, in the 1) RGB white background detection. In response to this, the operation (execution of step S4) is not performed until the generation of the color-corresponding white background information MS (steps S3-4-5-14 to 17) is started.
This is to prevent a region that cannot be determined as a white region from being erroneously determined as a white pixel (white block) by white pattern matching of the G image data after edge enhancement processing of the filter unit 321, which will be described later.

When a character on a light-colored ground is subjected to the edge emphasis filter section 321, the data around the character becomes a value (white) lower in level than the original image data (colored background). When white pattern matching is performed with the subsequent data, that is, white background information M corresponding to the density
If the white area is determined based only on the generation of S (steps S7-13-5-14-17), it is easy to erroneously determine the surroundings of the character on the color ground as a white background. In step S3-4-5-14 to 17), the white background information MS is set to the maximum value so that white pattern matching for determining a white pixel (white block), which will be described later, is applied to an area determined to be a white area. If it is not a white background in step 3, white background information MS which is one parameter for determining whether or not to apply white pattern matching by checking the white background conditions in detail in step S6 and subsequent steps.
Is adjusted, the erroneous determination of a white pixel (white block) in the later-described white pattern matching of the G image data after the edge enhancement processing of the filter unit 321 is prevented.

For example, when the possibility of a color pixel is high,
The white background information MS is lowered (step S13), and if there is a possibility of a color pixel, the white background information MS is held (no change) (steps S11 to S13) and mistaken for a white pixel (white block) by white pattern matching described later. This prevents the data around the character from becoming a value (white) lower in level than the original image data (colored background).

Where the character is dense, the white background information MS is updated and propagated by the above-described processing (steps S3-5, 6-10 and 14-17). Is reduced. Also, among characters such as complicated characters (for example, "letter"), 1) there is a case where white detection cannot be performed by RGB white background detection.
3) White is detected by the valley white pixel detection, and the white background information MS is held on the path in which the YES output of step S12 goes straight to step S5, and the white background information is kept in the white background tendency. Less likely.

Further, as mentioned above, when the target pixel is a color pixel whose periphery is surrounded by a white background, the above-mentioned 2) a color ground determination signal d = “1” (color ground) which is an output of the color ground detection. Becomes
Even when the line is complicated, a light color pixel can be detected as a color ground, and a threshold th for checking whether the periphery of the pixel of interest is white
By setting wc low (thwc = 20), it is possible to strictly check whether or not the periphery of the light color pixel (pixel of interest) has a white background and detect the light color pixel as a color background. The possibility that the inside is erroneously determined as a picture can be further reduced.

As described above, since a light color pixel can be more accurately detected as a color background, when a color background is detected, the process proceeds from step S6 to step S13 in FIG. In addition to the possibility that the threshold is determined, the binarized white determination signal referred to in step S7 is generated with respect to a threshold thwss (for example, 40) at the time of generating the white background determination signal referred to in step S3. When the threshold value thwsb (for example, 50) at the time is set to a large value and the color background is not determined (step S6: NO)
5, the probability of being regarded as white by the binarization unit 323a is increased, and the process proceeds from step S7 to step 10 in FIG.
To increase the possibility of determining a white area.

That is, in the above 1) RGB white background detection, the threshold value thwss = 40, a strict white determination with a low probability of white determination is performed, and if a white background is determined there, the processing from steps S3 to S4 in FIG. The possibility that the character background is determined to be white by raising the state variable MS is increased. When the white determination is not made in the strict white determination, a strict color determination with high reliability for detecting a light color pixel, which is a light color background, as a color background, that is, the 2) color background detection Referring to the result, when it is not determined that the color is a ground, the threshold th is again high, which is likely to be determined to be white.
wsb = 50, that is, a soft white judgment, that is,
23a, if it is determined to be white, the state variable M
S is raised to increase the possibility that the character background is determined to be white (steps S7 to S10). This processing (steps S6 to
10), when the background density unevenness is lighter than a light color pixel detected as a color background, for example, when there is unevenness in the ground of the original such as back reflection, the state variable is linked to the fine ground unevenness of the original. MS is suppressed from largely changing in a binary manner, and the determination of whether or not the pixel is a white pixel in the next white pattern matching unit 323d is suppressed from being finely changed in the scanning direction. As a result, when the background is a light color background, fine color omission (white background) does not appear in conjunction with fine background unevenness of the original such as back reflection.

{White Pattern Matching Unit 323d} Whether the background is white is determined by whether or not there are continuous white pixels in a block of 5 × 5 pixels centered on the target pixel. Therefore, for the target pixel, when the following expression is satisfied,
The target pixel is temporarily determined to be a white pixel, and white pattern matching is performed: (non-color pixel & (white background information MS ≧ thw1 (13)) & 2
(Non-color pixel & (white background information MS ≧ thw2 (1)) & valley white pixel & binarized white) Here, the pixel of interest to be checked whether this conditional expression is satisfied is the step shown in FIG. S5 and the target of the white propagation processing of 14 to 17, which have undergone the processing, and the "white background information MS" in the conditional expression is the white background information of the target pixel on which the check is performed after the white propagation processing. MS [I].
Here, MS [I] is the white background information after the white propagation processing, and I is the position of the pixel of interest to be checked in the main scanning direction x. Is different from the position in the main scanning direction x of the pixel of interest for which is calculated.

In the above conditional expression, “non-color pixel” is 2) the color ground determination signal d output from the color ground detection is “0”, and “binary white” is the binary That the binarized white determination signal of the section 323a is "1" (binary white); and
"Valley white pixel" means that the detection result of the above 3) valley white pixel detection is a valley white pixel, respectively, and # means logical sum (OR: or). In the white pattern matching, the output (white pixel or not) determined by the above-described conditional expression is compared with FIG.
Of the vertical / horizontal / diagonal continuity pattern PMPa to PMPd. White circles added to the patterns PMPa to PMPd mean white pixels. It does not matter whether the other blank pixels are white pixels.

The white pixel distribution of the 5 × 5 pixel matrix centered on the target pixel is represented by the patterns PMPa and PMP shown in FIG.
If it corresponds to b, PMPc or PMPd, it is determined that the target pixel is a white pattern pixel.

<Gray Judgment> {Gray Pixel Detection} A hue division of R, G, B, Y, M, C, and Bk is performed, and a pixel with low density is detected for each hue. Hue division is the same as color determination described later. Here, the G data after the filter is compared with thgr, and if either the data is larger than the G data or a color pixel is detected by the color pixel detection of RGB white extraction, the following operation is performed, and the following operation is performed. If the condition is satisfied, a gray pixel is determined. Here, the threshold value is changed for each color because the maximum density of each ink is different. 4.1) RY hue area boundary (ry) R-2 * G + B> 0 4.2) YG hue area boundary (yg) 11 * R-8 * G-3 * B> 0 4.3) GC hue area boundary (gc) 1 * R-5 * G + 4 * B <0 4.4) CB hue area boundary (cb) 8 * R-14 * G + 6 * B <0 4.5) B- M hue area boundary (bm) 9 * R-2 * G-7 * B <0 4.6) MR hue area boundary (mr) R + 5 * G-6 * B <0 4.8) Y pixel pixel determination (Gry) (It is a color pixel) & (ry == 1) & (yg == 0) & (Maximum RGB value <thmaxy) 4.9) G pixel determination (grg) (It is a color pixel) & ( yg == 1) & (gc == 0) & (maximum value of RGB <thmaxg) 4.10) C pixel determination (grc) (color pixel) & (gc == 1) & (cb == 0) ) & (Maximum RGB value <thmaxc) 4.11) B pixel determination (grb) (color pixel) & (cb == 1) & (bm == 0) & (maximum RGB value <thmaxb) 4 .12) M pixel determination (grm) (It is a color pixel) & (bm == 1) & (mr == 0) & RGB maximum value <thmaxm) 4.13) R pixel determination (grr) (Yes) & (mr = 1) & (ry == 0) & (maximum RGB value <thmaxr) 4.14) When not a color pixel (grbk) (Maximum RGB value <thmaxbk) 4.15) Gray pixel Judgment When any one of the conditions 4.8) to 4.15) is satisfied, a gray pixel is determined. Note that “==” is a notation in the C language.

This processing is performed by the gray pixel detection unit 32 shown in FIG.
3b-1. As described above, the RGB white extraction unit 32
3b is a gray pixel detector 323b-1, and a color pixel detector 3 is
23b-2, and an RGB white background detection section 323b-3, to which R, G, and B image data are input. The output of the gray pixel detection unit 323b-1 is input to the gray pattern matching unit 323h, and the pattern matching result of the gray pattern matching is output to the gray expansion unit 3.
23i, and after performing expansion processing, the determination unit 323
m. Further, the outputs of the color pixel detection unit 323b-2, the RGB white background detection unit 323b-3, and the binarization unit 323a are input to the white determination unit 323c.
The determination result of 3c is input to the white pattern matching unit 323d, and the pattern matching result is input to the white pattern correction unit 323j and the white correction unit 323g. The correction result of the white pattern correction unit 323h is further added to the white expansion unit 3.
After being processed by the 23k and the white contraction unit 323l, it is input to the determination unit 323m, and the processing result of the white correction unit 323g is directly input to the determination unit 323m. If contraction processing is performed before expansion processing by the gray expansion section 323i, isolated points can be removed. Also, a white pattern matching unit 323d, a white pattern correction unit 323j,
White expansion section 323k, white contraction section 323l, and white correction section 3
23g is a configuration for detecting a boundary area between white and non-white. The output of the white correction unit 323g indicates the line width, and the white contraction unit 3
The output of 231 indicates a white area, and the gray expansion section 323
The output of i indicates that the density is medium. Therefore, the judgment unit 3
In the case of 23m, these three outputs are prioritized and determined, and the determination result is output to the subsequent stage. In this case, priority 1 is line width information from the white correction unit 323g, priority 2 is medium density information from the gray expansion unit 323i, and priority 3 is white area information from the white contraction unit 323l. is there.

{Gray pattern matching section 323h}
The gray pattern matching unit 323h performs the following pattern matching on the assumption that D is a gray pixel and bk is darker than the gray pixel. Copy manuscript is thin 200
Since the line pattern is a line pattern of 300 yen, the following pattern is adopted so that the copy original is also detected in gray. Those that match the following pattern are gray pixels. FIGS. 22A and 22B show patterns at this time. FIG. 22A shows a pattern for 200 lines, and FIG. 22B shows a pattern for 300 lines.

(D15 & D25 & D35 & D32 & D45 & D38 &! BK41 & D42 &! BK43 &! BK44 & D55 &! BK46 &! BK47 & D48 &! BK49 D52 & D65 & D58 & D75 & D85 & D95) # (D05 & D15 & D25 & D31 & D33 & D35 & D37 & D38 & D41 &! BK42 & D43 &! BK44 & D45 &! BK46 & D47 &! BK48 & D48 && D51 & D53 & D55 & D57 & D58 & D65 & D75 & D85) {White pattern correction unit 323j} White pattern correction unit 323
j, isolated by white pixel pattern matching (1 × 1, 1
× 2, 2 × 1, 2 × 2, 1 × 3, 3 × 1 white pixels) are deactivated. As a result, isolated pixels are removed.

{White Expansion Section 323k} In the white expansion section 323k, the correction result of the white pixel pattern matching is calculated as 7 × 4
1 is ORed.

{White contraction section 323l} In the white contraction section 323l, 1 × 33 of the result of white expansion in the white expansion section 323h is obtained.
Is performed. By performing white expansion and white contraction, the inactive pixels existing in the expansion and small area with respect to the correction result in the white pixel pattern matching unit 323d are removed. This determination result includes a non-white background-side boundary region with respect to a white background and a boundary portion. In other words, the area is larger than the white background.

{White Correction Unit 323g} In the white correction unit 323g, each of the four corners of the 15 × 11 pixels centered on the target pixel marked with “X” in the block pattern BCP in FIG.
When one or more white candidate blocks exist in each of the × 4 pixel regions, white block correction data is given to the target block. As a result, an area surrounded by a white background is defined as a white area.

{Gray expansion section 323i} Gray expansion section 3
In 23i, an 11 × 11 OR process is performed on the result of the gray pattern matching. This results in a slightly larger area than the gray area.

{Determination Unit 323m} In the determination unit 323m,
The result of the white correction unit 323g is active, or the result of the white contraction unit 323l is active and the gray expansion unit 323 is active.
A white background is used when the result of i is inactive. This can be expressed by the following equation.

Result of White Correction # (Result of White Shrinkage &
! Here, in the result of the white correction, the area surrounded by the white background is definitely determined as the white area, and the result of the white contraction &! In the result of the gray expansion, the area around dark black characters is defined as a white area, and the area with low density is defined as a non-white area.

In FIG. 13, the black protrusions surrounded by circles Bp1 to Bp4 indicate the 15
When one or more white candidate blocks are present in each of the 6 × 4 pixel regions at the four corners of the × 11 pixel, the target block is replaced with a white block by white block correction for providing white block correction data to the target block. Making a black area surrounded by a white background, such as a black protrusion surrounded by circles Bp1 to Bp4, a white area lowers the possibility of determining that area as a picture part. The non-white area is determined to be a pattern by the comprehensive determination unit 326 described later, but the possibility of erroneously determining a black area surrounded by a white background as a pattern, such as a black protrusion surrounded by circles Bp1 to Bp4, is reduced. Further, the boundary between the black and white backgrounds is determined as a white area (character area) based on the result of the white contraction and the gray expansion. It is possible to determine an edge. In a portion with low density, character edge determination is not performed.

<Adjustment of Character / Photo Judgment Level> As described above, the white area extraction unit 323 uses the RG in the white judgment unit 323c.
White background information, which is a state variable indicating the degree of white corresponding to the white background determination signal, the color ground determination signal d and the valley white pixel determination signal of the B white extraction unit 323b, and the binary white determination signal of the binarization unit 323a. Generate MS. Then, the white pattern matching unit 323d temporarily determines whether or not the target pixel is a white pixel based on the color ground determination signal d, the white background information MS, the binary white determination signal, and the valley white pixel determination signal, and includes the target pixel. Whether or not a pixel is a white pixel is determined by white pixel distribution pattern matching for the pixel matrix. Using this result and the results of the black determination unit 323e and the black pattern matching unit 323f, the white correction unit 323g determines whether the target pixel is a boundary between a black background and a white background boundary (white region: character region). Note that the white region extraction unit 323 has the circuit configuration of FIG. 21 in the case of determining a gray pixel, but the white / black determination is processed by the circuit configuration of FIG.

The white background determination signal (see step S3 in FIG. 5) of the RGB white extraction unit 323b is used for the R, G,
If all the B image data is smaller than the threshold thwss = 40, it is “1” (white background). When the threshold value thwss is increased, the probability of determining a large value of the white background information MS increases, and the probability of extracting the “white region” (the boundary between the black background and the white background: a character region) increases (ie, the pattern region is extracted). The probability of doing so). The opposite is true if the threshold thwss is reduced.

The binarized white determination signal (see step S7 in FIG. 5) of the binarizing section 323a is "1" (if the edge enhancement output of the G image data of the filter section 321 is smaller than the threshold thwsb = 50). Binarized white). This threshold th
When wsb is increased, the probability of determining a large value of the white background information MS increases, and the probability of extracting the “white region” increases (ie, the probability of extracting the picture region decreases). The opposite is true if the threshold thwsb is reduced.

The image data of the “white area” is subjected to image processing for clearly representing the character image in a later process. Therefore, if the threshold values thwss and thwsb are increased, the image processing with high priority is applied to the character. Is done. The image data of the non-white area, that is, the picture (photograph) area is subjected to image processing for faithfully representing a picture or a picture in a subsequent process, and thus the threshold value th
When wss and thwsb are reduced, the pattern (photo)
Is subjected to image processing having a high priority.

When the color / ground determination signal d (see step S6 in FIG. 5) of the RGB white extraction unit 323b is "1" (color background), the white background information MS is lowered, and the "white area" is extracted. (That is, the probability of extracting a picture region increases). 2) Color ground determination signal d in color ground detection
B. generating process , C.I. When the threshold thwc (for example, 20) used in is reduced, peripheral pixels (（and □ in FIG. 11)
At the same time, ie, the probability that (exclusive NOR of b and c) = “1” increases, and the probability of obtaining the color ground determination signal d = “1” (color ground) increases,
The probability of extracting the “white area” is low (that is, the probability of extracting the picture area is high).

Therefore, in this embodiment, a menu display of an input mode by key input and a key image (a parameter designation key and an up / down key) on a menu screen displayed on the liquid crystal display are displayed on the operation / display unit OPB of FIG. The threshold value thws is adjusted by adjusting the “character / photo level” in the parameter adjustment performed by the operation
The s, thwsb, and thwc are adjusted as follows: Parameter Character-side adjustment value Standard Photo-side adjustment value thwc 26 24 22 20 18 16 14 That is, the standard value (default) of the parameter “character / photo level” adjusted and set by the operator on the operation / display unit OPB is “3”, and the default value is the character / photo level. Photo levels and thresholds thwss, thwsb and t
hwc and a conversion table indicating the relationship between the IPU3 and the IPU3 shown in FIG.
When the power is turned on at 00 and the CPU 357 initializes the IPU 300, the CPU 357 reads the default value of the character / photo level from the ROM 358, reads the corresponding threshold values thwss, thwsb, and thwc from the conversion table and reads Is written in the register addressed to each threshold value, and is used in the above-described processing in the white area extraction unit 323. After that, the character / photo level is adjusted by input from the operation board OPB, and the adjusted value A is given from the main controller 10 to the CPU 357.
The CPU 357 determines the parameter t corresponding to the adjusted value A.
The values of hwss, thwsb and thwc are stored in ROM
The data is read out from the conversion table 358 and written in the parameter address register of the RAM 356.

The threshold value is set to a standard value thwss = 40, thws
When b = 50 and thwc = 20, when the operator increases the value of “text / photo level” by i (for example, 1) by using the operation board OPB, “Up”
When the thresholds thwss, thwsb and thwc are 2i
(2) The value is set to the value changed in the character priority direction. Conversely, when the operator decreases the value of “text / photo level” by “i” (for example, 1), the threshold value thws
s, thwsb, and thwc are determined to be 2i (2) values changed in the photograph priority direction.

<Dot extraction> {Dot extraction unit 324} As shown in FIG. 23, the halftone extraction unit 324 includes a first halftone peak detection unit 324a, a second halftone peak detection unit 324b, and a third halftone dot. Peak detecting section 324c, first halftone area detecting section 324d, third halftone area detecting section 324
e and temporary storage means (temporary memory) 324f. The G image data is input to the first halftone peak detector 324a and the third halftone peak detector 324c,
The B image data is input to the dot peak detector 324b. The first halftone dot region detection unit 324d receives the detection results of the first halftone dot peak detection unit 324a and the second halftone dot peak detection unit 324b, and the second halftone dot region detection unit 324e inputs the third halftone dot detection unit 324e. The detection result of the peak detection unit 324c is input. The temporary memory 324f temporarily stores the detection results of the first and second halftone dot area detection units 324d and 324f. Note that the dot extraction unit 324 in FIG.
Corresponding to N.

The first halftone dot detecting section 324a uses the G image data to obtain a pixel forming a part of a halftone dot (halftone peak pixel) from pixel density information in a two-dimensional local area of a predetermined size. ). When the following two conditions are simultaneously satisfied for the local region, the central pixel of the region is detected as a halftone dot peak pixel. Condition 1: The density level of the central pixel is the maximum (peak) or the minimum (valley peak) in the local area. Condition 2: The absolute value of the difference between the average of the density level of the pixel pair and the density level of the central pixel is equal to or larger than the threshold Th for all pixel pairs having a point-symmetric relationship with the central pixel.

Referring to FIG. 14, the detection process of first halftone dot detecting section 324a will be specifically described. This is an example in which a mask of a 5 × 5 pixel matrix (M × M pixel matrix in general) is used as a local region. The code of each pixel of the 5 × 5 pixel matrix is represented by the pattern MPp in FIG.
, The density Lc of the center pixel c3, which is the target pixel, is maximum or minimum compared to the density L1 to L8 of the surrounding pixels, and abs (2Lc−L1−L8) ≧ Lth and abs ( When 2Lc−L2−L7) ≧ Lth and abs (2Lc−L3−L6) ≧ Lth and abs (2Lc−L4−L5) ≧ Lth, the center pixel (Lc) of the mask is detected as the halftone dot peak pixel. The abs function means taking an absolute value. Lth is a threshold value (fixed value).

Specifically, the surrounding pixels are pixels to which a square of the surrounding pixel distribution pattern MPa or MPb shown in FIG. 14 is added. Surrounding pixel distribution patterns MPa and MPb
When either of the above-described halftone peak pixel detection based on the above is detected as a halftone peak pixel, a detection signal indicating the halftone peak pixel is given to the target pixel (center pixel c3) at that time. The two patterns are used in order to widely correspond to the number of halftone dots.

The pattern MPa is as follows: L1 = b2, L2
= B3, L3 = b4, L4 = c2, L5 = c4,
L6 = d2, L7 = d3, and L8 = d4. Here, L1 = b2 means that the density of the pixel b2 is
This means that the value is L1 in the halftone peak pixel detection calculation described above.

The pattern MPb is as follows: L1 = b2, L2
= A3, L3 = b4, L4 = c1, L5 = c5,
L6 = d2, L7 = e3, and L8 = d4.

For copying, enlargement in the sub-scanning direction y,
Since the reduction is performed at low and high document scanning speeds of the scanner 200, the scanner 200 provides image data that has been enlarged or reduced in the sub-scanning direction y. Therefore, at the time of reduction, the patterns MPa and MPb described above are used instead of FIG.
The patterns MPc and MPd shown above are used. At the time of enlargement, patterns MPe and MPf shown in FIG. 14 are used. Note that, in the patterns MPe and MPf, pixels with a triangle mark may be added to the above-mentioned “surrounding pixels”.

The second halftone peak detecting section 324b detects halftone peaks using the B data.
This is the same as the first halftone dot detection unit 324a. The first halftone peak detector 324a uses the G image data and therefore responds to most colors, but does not respond to Y. Therefore, the second halftone peak detector 324c uses the B image data. It is an auxiliary object for detecting the halftone dot peak of Y.

The halftone dot region detection 324c is a two-dimensional image of a predetermined size having the peak and valley halftone dot pixels detected by either the first halftone peak pixel detection 324a or the second halftone peak pixel detection 324b. , And the sum of the peak pixels of the peaks and valleys is defined as the count value P of the small area. When the count value P is larger than the threshold value Pth, all the pixels of the small area (or only the central pixel of the small area in the case of processing in units of pixels) are determined to be a halftone dot area. The result of the judgment is temporary memory 3.
24f.

With reference to FIG. 24A, the detection process of the third halftone dot detecting section 324c will be specifically described. The detection processing of the third halftone dot peak detection unit 324c is intended to detect 100 lines or less and 65 lines (newspaper halftone dots) or more, and a 7 × 7 pixel matrix (M
This is an example employing a (× M pixel matrix) mask. If the pattern shown in FIG. 24C is used, the density Lc of the central pixel group that is the target pixel is the density group L of the surrounding pixels.
Abs (2Lc-L1-L8) ≥Lth and abs (2Lc-L2-L7) ≥Lth and abs (2Lc-L3-L6) ≥Lth and abs (2Lc −L4−L5) ≧ Lth, the center pixel (Lc) of the mask is detected as a halftone peak pixel. The abs function means taking an absolute value. Lth is a threshold value (fixed value).

More specifically, the surrounding pixels are pixels having a surrounding pixel distribution pattern shown in FIG. When either of the first and second halftone peak detectors 324a and 324b detects a halftone peak pixel based on the surrounding pixel distribution pattern, the pixel of interest at that time (center pixel d4)
Is supplied with a detection signal representing a halftone dot peak pixel. The two patterns are used in order to widely correspond to the halftone dot area ratio.

The density of Lc is determined as follows with reference to peripheral pixels. Lc = Min (d4, d3, d5, c4, e4) When Lc is the maximum value for the peripheral pixels, the pattern is as follows: L1 = Max (a1, a2, b1) L2 = Max (a3, a4, a5) L3 = Max (a6, a7, c7) L4 = Max (c1, d1, e1) L5 = Max (c7, d7, e7) L6 = Max (f1, g1, g2) L7 = Max (g3, g4, g5) ) L8 = Max (g6, g7, f7). Here, L1 = Max (a1, a
2, b1) means that the maximum value of the density of the pixels a1, a2, b1 is used as the value of L1 in the above-described halftone peak pixel detection calculation. What is Lc = Min (d4, d3, d5, c4, e4)?
Lc is the smallest of d4, d3, d5, c4, and e4.

Further, Lc = Max (d4, d3, d5,
c4, e4) When this Lc is the minimum value for the peripheral pixels, the pattern is: L1 = Min (a1, a2, b1) L2 = Min (a3, a4, a5) L3 = Max (a6, a7, c7) L4 = Max (c1, d1, e1) L5 = Max (c7, d7, e7) L6 = Max (f1, g1, g2) L7 = Max (g3, g4, g5) L8 = Max (g6, g7, f7) ).

In the case of copying, enlargement in the sub-scanning direction y,
Since the reduction is performed at low and high document scanning speeds of the scanner 200, the scanner 200 provides image data that has been enlarged or reduced in the sub-scanning direction y. Therefore, at the time of reduction, the pattern shown in FIG. 14B is used. At the time of enlargement, the pattern shown in FIG.

The arithmetic expression of the third halftone dot detecting section 324c does not operate on one pixel data, but refers to a target pixel by a plurality of pixels (calculation of min and max). This is because halftone dots having a low screen ruling have a large shading cycle. Therefore, if one pixel is determined, the influence of noise (garbage) is reduced by referring to neighboring pixels, and the amount of arithmetic operation is reduced. Since the arithmetic expressions can be used in common for the other blocks, it is easy to implement hardware.

The first halftone dot area detecting section 324d counts the peak pixels of the peaks and valleys detected by the first halftone dot detecting section 324a for each two-dimensional small area of a predetermined size. The sum of the peak pixels of the peaks and valleys is calculated as the count value P of the small area.
And When the count value P is larger than the threshold value Pth,
All pixels in the small area (or, in the case of pixel-based processing, only the center pixel of the small area) are determined as halftone areas. The result of the determination is stored in the temporary memory 324f.

If either the first halftone dot area detection section 324d or the second halftone dot area detection section 324e is a halftone area, the halftone / non-halftone area determination of the processed area near the small area of interest is performed. Threshold value Pth adaptively according to the result (peripheral feature information)
To change. In the present embodiment, two values TH1 and TH2 (where TH1> TH2) are set as the threshold value Pth.
Is prepared, and one of the values is selected according to the determination result of the processed area near the small area of interest stored in the temporary memory 324f. In other words, if the neighboring area is determined to be a non-dot area, there is a high possibility that the area is a line drawing area.
Is selected as the threshold value Pth. On the other hand, when it is determined that the neighboring area is a halftone area, it is highly possible that the neighborhood area is a halftone area, and therefore, TH2 under which the condition is relaxed is used as the threshold Pth. Note that TH1 is selected as the initial value of the threshold value Pth.

The distribution of the aforementioned small area is shown in AMP on FIG. Each of S1 to S4 of the small area distribution pattern AMP is a small area (block) having a size of, for example, 4 × 4 pixels, and S4 is a small area of interest, and S1, S2, and S3 are processed small areas. And S1, S2
When all of S3 and S3 are determined to be halftone areas, Th2 is used as the threshold value Pth for the determination of S4. When at least one of S1, S2 and S3 is determined to be a non-dot area, TH1 is selected as the threshold Pth. A halftone area detection signal ht of “1” is determined when the area is determined to be a halftone area, and “0” is determined when the area is determined to be a non-halftone area. However, this is only an example, and TH2 is selected when any one of S1, S2 and S3 is determined to be a halftone dot region, and TH1 is selected only when all of the small regions are determined to be non-halftone regions. May be selected. Further, the neighboring area to be referred to may be only S1 or only S2.

{Color determination section 325} When detecting a color (chromatic) pixel or a black (achromatic) pixel in a document, the relative reading deviation of R, G, and B is caused by sampling of each color image data. And exists for mechanical accuracy. This will be described with reference to FIG. FIG. 15A shows an image density signal. The black density signal is ideally black when the levels of the R, B, and G density signals match. However, the actual image data forms an image on the CCD 207 by the lens 206,
The image signal of the CCD 207 is digitized.
5 (b) is an ideal high / low waveform. However, since a general scanner uses a three-line CCD sensor, R, G, and B images of image data are not read simultaneously at the same time. Since they are arranged at intervals and cannot be read simultaneously in terms of time, a reading position shift will inevitably occur. For example, R and R representing the level change black shown in FIG.
The G and B color density signals are relatively shifted as shown in FIG. If this shift is large, a color shift appears at the periphery of the black area.

{Hue division unit 325a} Color judgment unit 325
Finds a chromatic area. Input data R,
G and B are converted into c, m, y and w (white) signals for color determination by the hue division unit 25a. As an example of hue division, the boundary of each color is obtained, and R, G,
B is the difference between the maximum value and the minimum value of each image data in RGB.
The difference was defined as follows. Here, R, G,
The B image data becomes black (darkens) as the number increases.

1) RY hue area boundary (ry) R-2 * G + B> 0 2) YG hue area boundary (yg) 11 * R-8 * G-3 * B> 0 3) GC Hue area boundary (gc) 1 * R-5 * G + 4 * B <04) CB hue area boundary (cb) 8 * R-14 * G + 6 * B <05) BM hue area boundary (bm) 9 * R-2 * G-7 * B <06 6) MR hue area boundary (mr) R + 5 * G-6 * B <07 7) w (white) pixel determination for color determination: (R <thwa) &
If (G <thwa) & (B <thwa), y = m =
Let c = 0. thwa is a threshold value.

8) Y pixel determination: (ry == 1) & (yg
== 0) & (RGB difference> thy), y = 1, m =
Let c = 0. thy is a threshold value.

9) G pixel determination: (yg == 1) & (gc
== 0) & (RGB difference> thg), c = y = 1,
Let m = 0. thg is a threshold value.

10) C pixel determination: (gc == 1) & (c
b == 0) & (RGB difference> thc), c = 1, m
= Y = 0. thc is a threshold value.

11) B pixel determination: (cb == 1) & (b
m == 0) & (RGB difference> thb), then m = c =
1, y = 0. thb is a threshold value.

12) M pixel determination: (bm == 1) & (m
r == 0) & (RGB difference> thm), m = 1, y
= C = 0. thm is a threshold value.

13) R pixel determination: (mr == 1) & (r
If y == 0) & (RGB difference> thr), y = m =
1, and c = 0. thr is a threshold value.

14) BK pixel determination: If none of the conditions 7) to 13), y = m = c = 1.

Further, the determination of the w pixel for color determination is performed. If the condition is (R <thw) & (G <thw) & (B <thw), the color pixel is set as w pixel and output as w. th
w is a threshold value. Here, the priorities of 7) to 14) are as follows:
Give priority to the smaller number. The aforementioned threshold values thwa and th
y, thm, thc, thr, thg, and thb are threshold values determined before copying (processing). The relationship between thw and thwa is thw> tha. The output signal is c,
m and y are 3-bit data of 1 bit each, and 1 bit of w of color pixel detection for color determination. Here, the reason why the threshold value is changed for each hue is that a threshold value corresponding to the hue region is determined for each hue region when the chromatic range is different. This hue division is an example, and any formula may be used.

Outputs c, m, y, w of hue division section 325a
Stores 5 lines in the line memories 325b to 325e,
It is input to the color pixel determination unit 325f.

{Color Pixel Determination Unit 325f} FIG. 6 shows the contents of the color pixel determination unit 325f. C, m for 5 lines
The data of y and w are stored in the pattern matching units 325f5 to 325f5.
325f7 and the counts 325f1 to 325f4. Here, the pattern matching unit 325f6 in the flow for obtaining the B / C signal will be described first.

{Pattern matching unit 325f6} When w pixel for color pixel exists, c = m = y = 0 of that pixel
To be corrected. With this correction, 5
The white level of the × 5 pixel matrix increases. Next, the pixel of interest is c, m, of the pixel determined by the hue division unit 325a.
y is all 1 (c = m = y = 1) or all 0 (c = m = y = 1)
m = y = 0) is determined to be a pixel (color pixel) other than the 5 ×
The determination is made by checking whether the 5-pixel matrix matches the next pattern. 1) Color pixel pattern group 1-1) Pattern 1-1 (pm1) D23 & D33 & D43 1-2) Pattern 1-2 (pm2) D32 & D33 & D34 1-3) Pattern 1-3 (pm3) D22 & D33 & D44 1-4) Pattern 1-4 (pm4) D24 & D33 & D42 The central pixel (pixel of interest) is D33. FIG. 16 shows these patterns pm1 to pm4. Open circles on these patterns indicate that at least one of c, m, and y is 1. The reason for employing the pattern matching is to prevent isolated points and the like from being picked up. Conversely, when detecting a small area color such as a halftone dot, the center pixel is 1 (c = m =
The determination may be made based on whether pixels (color pixels) are other than 0 (c = m = y = 0) or all of them are y = 0.

2) Pattern group for thin color line A color line surrounded by white is detected. The pattern used for this is shown in FIG. In FIG. 17, the pixels with white circles are:
c, m, and y are all 0 pixels. Data (c, c) of a 5 × 5 pixel matrix centered on the pixel of interest (center pixel)
m, y) are the patterns pw11a to pw of FIG.
When matching with any one of 14d, the target pixel (center pixel) at that time is regarded as a color line pixel. 2-1) Pattern 2-1 (pw11a to pw11d) ((D12 & D13 & D14) & (D42 & D43 & D44)) # ((D12 & D13 & D14) & (D52 & D53 & D54)) # ((D22 & D23 & D42) & (D42 & D43 & D44)) ((D22 & D22 & D) )) 2-2) Pattern 2-2 (pw12a to pw12d) ((D21 & D31 & D41) & (D24 & D34 & D44)) # ((D21 & D31 & D41) & (D25 & D35 & D45)) # ((D22 & D23 & D24) & (D24 & D34 & D44) & (D22 & D24 &&) (D25 & D35 & D45)) 2-3) Pattern 2-3 (pw13a to pw13d) ((D11 & D21 & D12) & (D35) D44 & D53)) # ((D11 & D21 & D12) & (D45 & D44 & D55)) # ((D13 & D22 & D31) & (D35 & D44 & D53)) # ((D13 & D22 & D31) & (D45 & D44 & D55)) 2-4) Pattern 2-4 (pw14a & Dw14 & Dw14 & Dw14 & Dw14) & (D41 & D51 & D52)) # ((D14 & D15 & D25) & (D41 & D51 & D52)) # ((D13 & D24 & D35) & (D31 & D42 & D53)) # (((D14 & D15 & D25) & (D31 & D42 & D53)) &lt; 3 &gt; However, pattern matching is performed. The pattern used for this is shown in FIG. In FIG. 18, the pixels with white circles are pixels in which c, m, and y are all 0. If the distribution of data (c, m, y) of the 5 × 5 pixel matrix centered on the target pixel (center pixel) matches any of the patterns pw21a to pw24d in FIG. ) Are considered as white area pixels. 3-1) Pattern 3-1 (pw21a to pw21d) (D21 & D31 & D41) # (D22 & D32 & D42) # (D24 & D34 & D44) # (D25 & D35 & D45) 3-2) Pattern 3-2 (pw22a to pw22d) (D12 & D13 & D14) & (D22 &# 24 &#) D42 & D43 & D44) # (D52 & D53 & D54) 3-3) Pattern 3-3 (pw23a to pw23d) (D52 & D51 & D41) # (D53 & D42 & D31) # (D35 & D24 & D13) # (D25 & D15 & D14) 3-4) &lt; D &gt;# (D53 & D44 & D35) # (D31 & D22 & D13) # (D21 & D11 & D12) 4) Color pixel candidate 2 If the pattern matching result extracted above matches the following pattern, the pixel of interest is set as color determination color pixel candidate 2: ((pm1 == 1) & ((pw11 == 1) # (pw21! =) 1))) # ((pm2 == 1) & ((pw12 == 1) # (pw22! = 1))) # ((pm3 == 1) & ((pw13 == 1) # (pw23! = 1))) # ((pm4 == 1) & ((pw14 == 1) # (pw24! = 1))) Here, (pm1 == 1) indicates that the data distribution centered on the pixel of interest is This means matching with the pattern pm1, and (pw11 == 1) means the pattern pw11
a to pw11d means that (pw21! = 1) is the pattern pw21a to pw11d
21d means matching. &
Indicates a logical product, and # indicates a logical sum. By this pattern matching, a color pixel surrounded by a white region is set as a color pixel candidate, and when there is a white region other than that, the pixel is not a color pixel. A match in color pixel pattern matching without a white region is a color pixel candidate.

{Counting unit 325f1} When a w pixel for color determination exists in a 5 × 5 pixel matrix centered on the pixel of interest, the c, m, y data determined by the hue dividing unit 325a for that pixel is used. Correct to c = m = y = 0. This correction increases the white level of the pixel matrix.
Then, 1 of c, m, y of each pixel in the pixel matrix is calculated.
The number of (c = 1, m = 1, y = 1) is counted. c,
If the difference between the maximum value and the minimum value of the count values for each of m and y is equal to or more than thcnt and the minimum value is less than thmin, the color pixel candidate 1 is determined. thcnt, thmi
n is a threshold value set before copying (processing). y, m,
The plane is expanded to c, the number is counted for each plane in the N × N matrix, and the minimum value is assumed to be black. As a result, even if reading of black pixels is omitted, correction can be performed. The chromatic pixel is determined based on the difference between the maximum value and the minimum value. Thus, the chromatic pixels are extracted by correcting the pixels from which the black pixels are not read. If there is a certain chromatic pixel in a 5 × 5 pixel matrix centered on the target pixel, the target pixel is determined to be a chromatic pixel.

{Color Pixel Determination Unit 325f8} Based on the outputs of the pattern matching unit 325f6 and the count unit 325f1, the color pixel determination unit 325f8 determines whether the pixel is a color pixel. If color pixel candidate 1 is color pixel candidate 2,
Let it be color pixel 1.

{Blocking Unit 325f9} The output of the color pixel determination unit 325f8 is blocked by the blocking unit 325f9. Blocking means that if there is one or more color pixels 1 in a matrix of 4 × 4 pixels, the 4 × 4 pixels
The entire pixel matrix is output as one color pixel block. The processing after the blocking unit 325f9 is
4 × 4 pixels are output as one block as a block.

{Isolated point removing unit 325f10} The isolated data is removed by the isolated point removing unit 325f10 as an isolated point if there is no color pixel block in a block adjacent to the target block.

{Expansion part 325f11} Isolated point removal part 32
The output of 5f10 is output to the color pixel 1 by the expansion unit 325f11.
If a block is present, it expands to a 5 × 5 block. The expansion is performed so that black character processing is not performed around color pixels. Here, the output B / C signal outputs L (chromatic) when the color pixel is one block, and outputs H (achromatic) otherwise.

{Counting unit 325f2} When a w pixel for color determination exists in a 5 × 5 pixel matrix centered on the pixel of interest, the c, m, y data determined by the hue dividing unit 325a for that pixel is converted to c = M = y = 0. This correction increases the white level of the pixel matrix.
Then, the number of 1 of c, m, y (c = 1, m = 1, y = 1) of each pixel in the pixel matrix is counted.
The difference between the maximum value and the minimum value of the count value for each of c, m, and y is greater than or equal to thacnt and the minimum value is tha.
If it is less than min, the target pixel is set as color pixel candidate 1. t
“hacnt” and “thamine” are thresholds set before copying (processing).

{Color Pixel Determination Unit 325f12} Based on the outputs of the pattern matching unit 325f6 and the count unit 325f2, the color pixel determination unit 325f12 determines whether the pixel is a color pixel. If the color pixel candidate is 1 and the color pixel candidate is 2, color pixel 2 is set.

{Blocking Unit 325f13} The output of the color pixel determination unit 325f12 is blocked by the blocking unit 325f13. That is, if there is one or more color pixels 2 in a 4 × 4 pixel matrix, the entire 4 × 4 pixel matrix is output as two color pixel blocks. In the processing after the blocking unit 325f13, 4 × 4 pixels are output as one block as one block.

{Density part 325f14} In order to remove an isolated block, if there are three or more active conditions (two color pixel blocks) in a 3 × 3 block and the target block is active (color pixel), the target block Is an active block (two blocks of color pixels).

{Counting section 325f3} The number of 1 (c = 1, m = 1, y = 1) of c, m, y of each pixel in the 5 × 5 pixel matrix centering on the target pixel is counted.
If the difference between the maximum value and the minimum value of the count value for each of c, m, and y is equal to or greater than tha1cnt and the minimum value of c, m, and y is less than tha1min,
Let it be color pixel candidate 3. tha1cnt, tha1min
Is a threshold value set before copying (processing).

{Pattern Matching Unit 325f5} Whether the pixel (c, m, y) determined by color pixel detection is a color pixel is determined by pattern matching using a 5 × 5 pixel matrix. The pattern is a pattern matching unit 325f6
Is the same as The pixel matched by the pattern matching is set as a color pixel candidate 4.

{Color pixel determination unit 325f15} If color pixel candidate 3 is color pixel candidate 4, color pixel 3 is determined.

{Blocking Unit 325f16} The output of the color pixel determination unit 325f15 is blocked by the blocking unit 325f16. That is, if there is one or more color pixels 3 in a 4 × 4 pixel matrix, the 4 × 4 pixels
The entire pixel matrix is output as three blocks of color pixels. The processing after the blocking unit 325f16 is 4 ×
4 is output as one block.

{Density part 325f17} In order to remove an isolated block, if there are three or more active conditions (three color pixel blocks) in a 3 × 3 block and the target block is active (color pixel 3), Let the block be an active block (three color pixel blocks).

{Count section 325f4} 1 of c, m, y determined by the hue division section 325a for each pixel in the 5 × 5 pixel matrix centering on the target pixel (c = 1, m = 1,
The number of y = 1) is counted. If the minimum value of each of the count values of c, m, and y is equal to or larger than thabk, the pixel of interest is set as black pixel candidate 1. thabk is a threshold value set before copying (processing).

{Pattern matching unit 325f7} In a 5 × 5 pixel matrix centered on the target pixel, c =
Pattern matching is performed for the pixel of m = y = 1. 1-1) Pattern 1-1 (pm1) D23 & D33 & D43 1-2) Pattern 1-2 (pm2) D32 & D33 & d34 1-3) Pattern 1-3 (pm3) D22 & D33 & D44 1-4) Pattern 1-4 (pm4) D42 & D33 & D24 The pattern shown in FIG. 16 is a pixel indicated by a circle in the figure is a pixel of c = m = y = 1. When it matches any of these patterns, the pixel of interest is set as black pixel candidate 2.

{Achromatic determination section 325f18}
If it is the black pixel candidate 1 and the black pixel candidate 2, it is determined as a black pixel.

{Blocking Unit 325f19} The black pixel output is blocked by the blocking unit 325f19.
In this case, if there is one or more black pixels in a 4 × 4 pixel matrix, the entire 4 × 4 pixel matrix is output as a black pixel block.
In the processing after the blocking unit 325f19, 4 × 4 pixels are output as one block as one block.

{Expansion unit 325f20} In the 3 × 3 block matrix, if the target block is active (black pixel block) and its surrounding pixels are non-active (non-black pixels), the target block is non-active (non-black pixel). Pixel block).

{Total color pixel determination unit 325f21} If the target block is determined to be active (color pixel 2) by color pixel determination unit 325f12 and not determined to be active (black pixel) by achromatic determination unit 325f18, The block is determined to be a color (color block). The color is also determined when the color pixel determination unit 325f15 is active (color pixel).

{Expansion Unit 325f22} In order for the overall color pixel determination unit 325f21 to consider small characters to be continuous with respect to the block determined to be a color, even one block in a 9 × 9 block matrix centered on the target block. If there is an active block, the block of interest is set as the active block. Here, the reason for the large expansion is to fill the gap between the characters.

{Continuous counting section 325f23} The continuous counting section 325f23 determines whether the document is a color document or a monochrome document by checking the continuity of the color pixel blocks. Expansion part 325
It is determined whether the original is a color original by counting the number of consecutive color pixels in the output data (color pixel block) of f22.

FIG. 7 shows the contents of this determination processing. When the target pixel is in the color pixel block, the upper left, upper,
The number of continuous color pixels of the target pixel is calculated with reference to the number of continuous color pixels of the upper right pixel and the left pixel (steps S21 and S2).
6). Here, if the target pixel is, for example, the c3 pixel of the 5 × 5 pixel distribution pattern MPp in FIG. 11, the upper left, upper, upper right, and left pixels are b2, b3, b4, and c2 pixels, respectively. When the target pixel is not in the color pixel block, the color pixel continuation number of 0 is given to it (steps S21 to S27).

When the target pixel is in the color pixel block,
First, the number of consecutive color pixels of the upper pixel (b3) of the target pixel (c3) is checked (step S22). If it is 0, 1 is set to the reference value A as the number of consecutive color pixels of the upper right pixel (b4). The added value is given (step S24). If the number of continuous color pixels of the upper pixel (b3) is 0, the upper right pixel (b4) is added to the reference value A.
Is given (step S23). Next, a value obtained by adding 1 to the number of continuous color pixels of the upper left pixel (b2) is given to the reference value B, a value obtained by adding 1 to the number of continuous color pixels of the upper pixel (b3) is given to the reference value C, The reference value D includes the left pixel (c
A value obtained by adding 1 to the color pixel continuation number of 2) is given (step S25). Then, the highest value among the reference values A, B, C, and D is set as the color pixel continuation number of the pixel of interest (c3) (step S26).

When the number of continuous color pixels is given to the pixel of interest (c3) as described above, the number of continuous color pixels becomes the set value THACS.
It is checked whether it is the above (step S28), and TH
If it is equal to or greater than ACS, it is determined that the original is a color original (step S29), and the process of the continuous counting unit 325f23 ends there. If the number of continuous color pixels is less than the set value THACS, the target pixel is updated to the next pixel in the scanning direction x, y, and the above-described processing is repeated. As a result of performing the above-described processing on the entire original, the number of continuous color pixels is set to the set value
If it is less than (Steps S30 to S34), it is determined that the document is a monochrome image.

The above-described number of continuous color pixels is substantially the sum of a vertical colored line segment and a horizontal colored line segment. The upper right color pixel continuation number is different from the others in order to prevent double counting. FIG. 19 shows specific data of the color pixel continuation number. FIG.
Is a color pixel, and the number is the number of continuous color pixels given to the pixel. A block formed by a series of small squares containing numbers is a color pixel group, and even if the number of color pixel continuations in any color pixel group on the same document is one, the set value T
If HACS or higher, it is determined whether the original is a color original in color or black and white (steps S28 and S28).
9).

The color pixel determination units 1 to 3 (325f8-325)
The reason for f15) is to increase the accuracy of determining whether a document is a color document or a monochrome document. The color pixel determination for black character processing is local and inconspicuous even if an erroneous determination is made. However, the determination as to whether the document is a color document or a black-and-white document affects the entire document if a wrong determination is made. Therefore, the counting unit 325f1
-F4 was made independent. Normally, the hue division unit 325a
However, it is not preferable to make the hue division unit 325a independent since the memory of the pattern matching units 325f5-f7 increases. Counting part 325f1
By changing the parameters of the color pixels (color pixels 1-3) with the parameter of -f4 (color pixel candidates 1, 3 and black pixel candidate 1), the increase in the memory amount is reduced. The color pixel determination units 2 and 3 (325f12 and 325f15) are provided to detect a low-density color such as a yellow color of a highlighter pen, and are further provided with an achromatic determination (black pixel determination) unit 32.
The reason why 5f18 is provided is to compensate for erroneous detection when the density is lowered. There is no problem if a light color such as a highlighter pen is corrected with black data in a certain width. When extracting a plurality of color pixels, only the level of w (white) is changed, so there is no need to have two memories for detecting color pixels, and a capacity of one line plus one line is possible. is there.

Since the count value is counted by referring to the count data of the previous line and the count data of the current line in the continuous count unit 325f23, the continuation of the peripheral pixels can be counted accurately without fail. It is possible to count the sequence of color pixels. In the present embodiment, R,
The hue was determined for the G and B image data.
The hue determination is not limited to the G and B image data, but may be performed on the luminance / color difference (for example, Lab).
Easy.

{Comprehensive decision section 326} Comprehensive decision section 326
As shown in FIG. 4, the character determination unit 326a, the expansion processing unit 326b, the character absence determination unit 326c, and the decoding unit 3
26d.

{Character determination section 326a} Character determination section 326
In a, the result of the edge extraction unit 322 is an edge, and the result of the halftone dot extraction unit 324 is a white area extraction unit 323 without a halftone dot.
If there is a white area in the result, it is determined to be a character edge.
If not, it is determined as a non-character edge (inside a picture or a character), and the result is output to the expansion processing unit 326b as shown in FIG.

{Expansion Processing Unit 326b} Expansion Processing Unit 326
In b, the result of the character determination unit 326a is subjected to OR processing of 8 × 8 blocks, and thereafter, AND processing of 3 × 3 blocks is performed to perform expansion processing of 4 blocks. That is, if any of the 8 × 8 blocks around the block of interest is a character edge, it is assumed that the block of interest is also a character edge block, and 3
If all of the × 3 blocks are character edges, the target block is determined as a character edge, and the target block and three blocks adjacent thereto are regarded as character edges, for a total of four blocks. O
The reason why the AND processing is performed after the R processing is that, in the case of a black character in particular, if a small non-black character area exists around the black character area, a feeling of incongruity may be felt due to a difference in processing. It is. To prevent this, the non-black character area is enlarged by OR processing. AND
The processing is performed to obtain a desired expansion amount.

By the way, since a color copying machine scans four times to make one copy, the character judgment result slightly differs for each scan. In particular, if a non-black character determination is made during black image formation and a black character determination is made at times other than black image formation, the black character area becomes thin.
Sometimes 8 × 8 blocks are ORed and then 3 × 8
When an image other than bk is formed by performing AND processing of three blocks, 5
OR processing is performed on the × 5 blocks, and then, AND processing is performed on the 1 × 1 blocks. Note that performing 1 × 1 AND processing is the same as not performing any processing because the result is the same as before the processing. The result of the dilation processing is output as a character edge signal by the decoding unit 326d.
Output to

By performing the expansion processing in this manner, the character area is not thinned due to a different separation result. Although the middle part of the character may be darkened by the dilation processing, the density is saturated when the character is light relative to the edge of the character, so that there is no sense of incongruity.

FIG. 20 is a schematic enlarged view showing the overlap of color coloring materials by color copying. (D) of FIG.
All four colors show ideal cases in which black character processing has been performed. FIG. 20E shows a case where black characters are processed for all four colors, and only bk is not corrected, but correction is applied to other than bk, and the color becomes lighter. (F) of FIG. 20 shows a preferable case in which black character processing is performed only on bk according to the present embodiment, and (g) of FIG.
This embodiment shows a preferable case in which black character processing is performed only on bk, correction is not performed on only bk, and correction is performed on other than bk.

FIG. 20A shows an ideal case where the amount of expansion is the same and black character processing is performed. FIG. 20B shows the case where the expansion amount is the same and the black character processing is performed, and the printing position is shifted (whitened out). FIG. 20C shows a case where the expansion amount of bk is large, and a case where the printing position is shifted due to black character processing according to the present embodiment.

{Decoding Unit 326d} Decoding Unit 326
The C / P signal finally output by d is as follows: C / P signal Character edge signal 0 None 1 Yes Character edge area Also from the color determination unit 325, as shown in FIGS. A B / C signal is output.

Next, FIG. 3 will be referred to again. Document recognition unit 3
The C / P signal and the B / C signal generated by 20 are
Filter section 330, color correction section 340, scaling section 350, interface section 352, UCR section 360, CMYBk
The filter unit 370, the CMYBkγ correction unit 380, and the gradation processing unit 390 are provided to the cascade in synchronization with the image data.

The RGB filter section 330 is a filter for performing MTF correction on RGB data, and is composed of a coefficient matrix corresponding to an N × N pixel matrix and logic for obtaining a weighted average value by multiplying each coefficient by each image data. ing. C / P signal representing 3 (character edge area)
, Use a coefficient matrix for sharpening processing,
Represents 0 or 1 (characterless area or picture area)
, The coefficient matrix for the smoothing process is used,
The weighted average value is derived and output to the color correction unit 340. The color correction unit 340 converts the R, G, and B data into C, M, and Y data by a primary masking process or the like. Magnification section 350
Performs enlargement / reduction or equal magnification processing in the main scanning direction x on image data.

The UCR section 360 is for improving the color reproduction of the image data, and converts the common part of the C, M, and Y data input from the color correction section 340 into a UCR (additive color removal).
Process to generate Bk data and output C, M, Y, Bk data. Here, when the C / P signal is other than 1 (character edge area) (when the area is a character area or a picture area)
Performs skeleton black processing. C / P signal is 3
In the case of (character edge area), full black processing is performed.
When the C / P signal is 1 (character edge area) and the B / C signal is H (achromatic area), C, M, and Y data are erased. This is because black characters are represented only by black components.

The output image signal IM of the UCR unit 360 is
G is one color of C, M, Y, and Bk at the temporary point, and is a one-color output of the frame sequence. That is, full-color (four-color) data is generated by reading the document four times. In the case of black and white copying, only one image reading is required since only one Bk image formation is required. If there is a mechanism for determining whether a document is a color document or a black-and-white document, the number of readings according to the document is sufficient, so that the operator does not need to determine whether the document is a color document or a black-and-white document and copy the document. In the present embodiment, the B / C signal is a signal referred to when determining whether the document is a color document or a monochrome document. When the B / C signal is H (achromatic region) over the entire surface of the document, the main controller 10 determines that the document is a monochrome document.

The CMYBk filter section 370 performs N × N × F filter processing according to the frequency characteristics of the color printer 400 and the C / P signal.
Smoothing or sharpening processing is performed using N spatial filters.
The CMYBkγ correction unit 380 changes and processes the γ curve according to the frequency characteristics of the color printer 400 and the C / P signal. When the C / P signal is 0 (picture area), a γ curve that faithfully reproduces an image is used, and when the C / P signal is 3 (character edge area), the γ curve stands to enhance the contrast.

The gradation processing unit 390 is provided in the color printer 40.
Quantization such as dither processing and error diffusion processing is performed according to the gradation characteristic of 0 and the C / P signal. At the time of Bk image formation, when the C / P signal is 0 (pattern area), a process emphasizing gradation is performed, and at other times, a process emphasizing resolution is performed. At the time of image formation other than Bk, when the C / P signal is 0 (pattern area), the processing that emphasizes gradation is performed, and at other times, the processing that emphasizes resolution is performed. The image data having undergone the above processing is transmitted from the video control 359 having a buffer memory to the color printer 400.
At the same time as the image data writing operation.

In the picture processing (C / P signal = 0), the IPU 300 performs smoothing processing in the RGB filter section 330, performs skeleton black processing in the UCR section 360, and performs linear processing in the CMYBkγ correction section 380. Gradation)
Is selected, and the CMYBk filter unit 37 is selected.
The 0 and gradation processing section 390 performs processing that emphasizes gradation.

On the other hand, character processing (C / P signal = 1 and B / C
When (signal = L), the RGB filter 330 performs edge enhancement, the UCR 360 performs full black processing, the CMYBkγ correction 380 selects a curve emphasizing contrast, and selects the CMYBk filter 370 and the gradation. The processing unit 390 performs processing emphasizing resolution.

In addition, black character processing (B / C signal when C / P signal = 1)
As C signal = H), C, M, and Y data are not printed during C, M, and Y image formation except for Bk. This is to prevent the surroundings of the black characters from being colored due to misalignment. In addition, the RGB filter unit 33 of the Bk data at this time.
A value of 0 may be sharpened by performing edge enhancement more strongly than in the case of a color character.

As described above, the IPU 300 performs four types of processing, ie, a pattern, a character edge, a character on a pattern, and a character middle processing.

[0178]

As described above, according to the first aspect of the present invention, the medium density detecting means for detecting the medium density area of the image represented by the image data, and the image data detected by the medium density detecting means are expanded. Since the image processing apparatus includes the expansion unit and the non-character edge determination unit that determines a region expanded by the expansion unit as a non-character edge region of the image, the region expanded by the expansion unit may not be determined as a character. . As a result, the processing is switched only in the medium density area, and the same image processing can be performed in the dark line area regardless of the line width. Because of this, since the density of general printed matter is high, when an image is viewed in a picture area or a non-picture area, it can be processed well as a non-picture area regardless of the line width. It is possible to provide an image processing apparatus in which an image does not look unnatural at the boundary portion.

According to the second aspect of the present invention, since the expansion means performs the expansion processing after performing the contraction processing, the isolated point can be removed, and the same effect as that of the first aspect can be obtained. Play.

According to the third aspect of the present invention, since the detection of the medium density area is performed by pattern matching with the line pattern, the medium density pattern area in the copy original of the line pattern is definitely determined as the picture area. It is possible to do.

According to the fourth aspect of the present invention, since edge detecting means for detecting an edge is further provided, a line having a high density is determined as a character irrespective of the line width, and a character having only a thin line is determined as to a medium density. And processing of dark characters is not mixed. As a result, the density of the general original is high, and the thin line is excluded from the pattern matching. Therefore, only the thin line needs to be subjected to character determination, and the processing is facilitated.

According to the fifth aspect of the present invention, the image processing apparatus according to any one of the first to fourth aspects, and the image data generated by reading the original image after color separation is transmitted to the image processing apparatus. Since the image reading apparatus is configured to include the color scanner for inputting, it is possible to provide an image reading apparatus having the effects of the first to fourth aspects.

According to a sixth aspect of the present invention, an image is formed based on the image processing apparatus according to any one of the first to fourth aspects and image data output from the image processing apparatus. Since the image reading apparatus includes a color printer that forms the formed image on a sheet and outputs the image, an image forming apparatus having the effects of the first to fourth aspects of the invention can be provided.

According to the seventh aspect of the present invention, the image processing apparatus according to any one of the first to fourth aspects, and the image data generated by reading the original image by color separation is sent to the image processing apparatus. A color copying apparatus comprises a color scanner for inputting, and a color printer for forming an image based on image data output from the image processing apparatus, forming the formed image on paper, and outputting the image. Therefore, it is possible to provide a color copying apparatus having the effects of the first to fourth aspects of the present invention.

According to the eighth aspect of the present invention, there is further provided a printer controller for analyzing an external print instruction command and printing out external image information by the color printer. It can be used as an external input printer such as a printer.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of a digital full-color copying machine including an image processing apparatus according to an embodiment.

FIG. 2 is a block diagram showing an outline of an electric system of the digital full-color copying machine shown in FIG.

FIG. 3 is a block diagram illustrating a configuration of an image processing unit (IPU) in FIG. 2;

FIG. 4 is a functional block diagram illustrating a configuration of a document recognition unit in FIG. 3;

FIG. 5 is a flowchart illustrating a processing procedure for updating state variables MS and SS [I] used for white determination.

FIG. 6 is a functional block diagram showing the contents of a color pixel determination unit in FIG. 3;

FIG. 7 is a flowchart showing a processing procedure of a determination process for determining whether a document is a color document or a monochrome document in the continuous counting unit in FIG.

FIG. 8 is a diagram showing a 600 dpi line pattern compared with a 400 dpi line pattern.

FIG. 9 is a diagram showing a pattern of a 3 × 3 pixel matrix used for pattern matching performed by a black pixel continuation detection unit and a white pixel continuation detection unit in FIG. 4;

10 is a diagram illustrating an example of a pattern used for separating a white background by an RGB white background detection unit in FIG. 4;

FIG. 11 is a diagram showing an example of a pattern used when detecting a color ground in FIG. 4;

FIG. 12 is a diagram schematically showing a current line (line of interest) of the line memory LMP.

FIG. 13 is a diagram showing an area for explaining a process of a white area extracting unit in FIG. 4;

FIG. 14 is a diagram for describing a detection process of a first halftone peak detection unit of the halftone dot extraction unit in FIG. 4;

FIG. 15 is a diagram for explaining processing of a color determination unit in FIG. 4;

FIG. 16 is a diagram for explaining pattern matching in color pixel determination in FIG. 6;

FIG. 17 is a diagram showing a color fine line pattern for detecting a color line surrounded by white.

FIG. 18 is a diagram showing a white area pattern for performing pattern matching where c, m, and y are all 0.

FIG. 19 is a diagram showing specific data of the color pixel continuation number.

FIG. 20 is a diagram schematically showing an enlarged overlap of color coloring materials by color copying.

FIG. 21 is a block diagram illustrating details of a white region extraction unit in FIG. 4;

FIG. 22 is a diagram showing an example of pattern matching with a line pattern for detecting a gray pixel;
Is a pattern for 200 lines, and (b) is a pattern for 300 lines.

FIG. 23 is a block diagram showing details of a halftone dot extraction unit in FIG. 4;

FIG. 24 is a diagram for describing a detection process of a third halftone dot detection unit.

FIG. 25 is an explanatory diagram showing processing contents of a comprehensive judgment unit in FIG. 4;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Main controller 12 Scanner controller 16 Printer controller 17 FAX controller 18 Parallel I / F 19 LAN control 20 Communication controller 40 Image processing unit (IPU) 200 Color image reading device (scanner) 300 Image processing unit (IPU) 320 Document recognition unit 321 Filter section 322 Edge extraction section 323 White area extraction section 323a Binarization section 323b RGB white extraction section 323b-1 Gray pixel detection section 323b-2 Color pixel detection section 323b-3 RGB white background detection section 323c White determination section 323d White pattern matching Unit 323g white correction unit 323h gray pattern matching unit 323i gray expansion unit 323j white pattern correction unit 323k white expansion unit 323l white contraction unit 323m determination unit 3 4 Halftone dot extraction unit 324a First halftone peak detection unit 324b Second halftone peak detection unit 324c Third halftone peak detection unit 324d First halftone region detection unit 324e Second halftone region detection unit 324f Temporary memory 325 colors Judgment unit 326 Overall judgment 326a Character judgment unit 326b Expansion processing unit 330 RGB filter 400 Color image recording device (color printer)

────────────────────────────────────────────────── ─── of the front page continued (51) Int.Cl. ⁷ identification mark FI theme Court Bu (reference) H04N 1/60 H04N 1/40 D F term (reference) 2H030 AD11 BB02 BB24 BB42 BB63 5C077 LL19 MP02 MP05 MP07 PP27 PP32 PP52 PP53 PP55 PP68 PQ17 PQ20 SS01 SS02 TT02 TT06 5C079 HB01 HB03 LA03 LA06 MA04 MA11 NA06 PA02 PA03 5L096 BA17 EA02 FA06 FA14 FA44 GA55 JA03

Claims (8)

[Claims] 1. An image processing apparatus that performs predetermined processing on input image data and outputs the processed image data, wherein: a medium density detection unit that detects a medium density region represented by the image data; An image processing apparatus comprising: an expansion unit that performs expansion processing on the image data detected in step 1; and a non-character edge determination unit that determines a region expanded by the expansion unit as a non-character edge region of the image. . 2. The method according to claim 1, wherein the expansion unit performs a contraction process.
The image processing apparatus according to claim 1, wherein the expansion processing is performed. 3. The image processing apparatus according to claim 1, wherein said medium density detecting means detects the medium density area by pattern matching with a line pattern. 4. The image processing apparatus according to claim 1, further comprising edge detection means for detecting an edge. 5. An image processing apparatus according to claim 1, further comprising: reading means for inputting, to the image processing apparatus, image data generated by reading a document image by color separation. An image reading apparatus, comprising: 6. An image processing apparatus according to claim 1, wherein an image is formed based on image data output from the image processing apparatus, and the formed image is formed on a sheet. And an image output means for outputting an image. 7. An image processing apparatus according to claim 1, wherein said image processing apparatus reads color-separated document images and inputs the generated image data to said image processing apparatus. An image output unit that forms an image based on image data output from a processing device, forms the formed image on paper, and outputs the image. 8. A color copying apparatus according to claim 7, further comprising control means for analyzing an external print instruction command and printing out image information input from the outside by said image output means. apparatus.